Hugging Face's logo

Datasets:
stpete2
/
ipynb

License:
Dataset card Files
xet
Community
stpete2 commited on
Commit
4cbf937
·
verified ·
1 Parent(s): bb8ce19

Upload imc2025-only-colmap-09.ipynb

Browse files
Files changed (1) hide show
  1. imc2025-only-colmap-09.ipynb +1361 -0
imc2025-only-colmap-09.ipynb ADDED
@@ -0,0 +1,1361 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "markdown",
5
+ "id": "056398df",
6
+ "metadata": {
7
+ "papermill": {
8
+ "duration": 0.00261,
9
+ "end_time": "2026-01-12T15:57:43.783277",
10
+ "exception": false,
11
+ "start_time": "2026-01-12T15:57:43.780667",
12
+ "status": "completed"
13
+ },
14
+ "tags": []
15
+ },
16
+ "source": []
17
+ },
18
+ {
19
+ "attachments": {},
20
+ "cell_type": "markdown",
21
+ "id": "f300e4a7",
22
+ "metadata": {
23
+ "papermill": {
24
+ "duration": 0.0017,
25
+ "end_time": "2026-01-12T15:57:43.787188",
26
+ "exception": false,
27
+ "start_time": "2026-01-12T15:57:43.785488",
28
+ "status": "completed"
29
+ },
30
+ "tags": []
31
+ },
32
+ "source": [
33
+ "# **IMC2025 Only Colmap 09**\n",
34
+ "### **image matching and 3D reconstruction**\n",
35
+ "\n",
36
+ "**INTERNET ON**"
37
+ ]
38
+ },
39
+ {
40
+ "cell_type": "code",
41
+ "execution_count": 1,
42
+ "id": "f268b82a",
43
+ "metadata": {
44
+ "execution": {
45
+ "iopub.execute_input": "2026-01-12T15:57:43.792138Z",
46
+ "iopub.status.busy": "2026-01-12T15:57:43.791847Z",
47
+ "iopub.status.idle": "2026-01-12T15:57:49.822499Z",
48
+ "shell.execute_reply": "2026-01-12T15:57:49.821465Z"
49
+ },
50
+ "papermill": {
51
+ "duration": 6.03517,
52
+ "end_time": "2026-01-12T15:57:49.824191",
53
+ "exception": false,
54
+ "start_time": "2026-01-12T15:57:43.789021",
55
+ "status": "completed"
56
+ },
57
+ "tags": []
58
+ },
59
+ "outputs": [
60
+ {
61
+ "name": "stdout",
62
+ "output_type": "stream",
63
+ "text": [
64
+ "Collecting pycolmap\r\n",
65
+ " Downloading pycolmap-3.13.0-cp311-cp311-manylinux_2_28_x86_64.whl.metadata (10 kB)\r\n",
66
+ "Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (from pycolmap) (1.26.4)\r\n",
67
+ "Requirement already satisfied: mkl_fft in /usr/local/lib/python3.11/dist-packages (from numpy->pycolmap) (1.3.8)\r\n",
68
+ "Requirement already satisfied: mkl_random in /usr/local/lib/python3.11/dist-packages (from numpy->pycolmap) (1.2.4)\r\n",
69
+ "Requirement already satisfied: mkl_umath in /usr/local/lib/python3.11/dist-packages (from numpy->pycolmap) (0.1.1)\r\n",
70
+ "Requirement already satisfied: mkl in /usr/local/lib/python3.11/dist-packages (from numpy->pycolmap) (2025.3.0)\r\n",
71
+ "Requirement already satisfied: tbb4py in /usr/local/lib/python3.11/dist-packages (from numpy->pycolmap) (2022.3.0)\r\n",
72
+ "Requirement already satisfied: mkl-service in /usr/local/lib/python3.11/dist-packages (from numpy->pycolmap) (2.4.1)\r\n",
73
+ "Requirement already satisfied: onemkl-license==2025.3.0 in /usr/local/lib/python3.11/dist-packages (from mkl->numpy->pycolmap) (2025.3.0)\r\n",
74
+ "Requirement already satisfied: intel-openmp<2026,>=2024 in /usr/local/lib/python3.11/dist-packages (from mkl->numpy->pycolmap) (2024.2.0)\r\n",
75
+ "Requirement already satisfied: tbb==2022.* in /usr/local/lib/python3.11/dist-packages (from mkl->numpy->pycolmap) (2022.3.0)\r\n",
76
+ "Requirement already satisfied: tcmlib==1.* in /usr/local/lib/python3.11/dist-packages (from tbb==2022.*->mkl->numpy->pycolmap) (1.4.0)\r\n",
77
+ "Requirement already satisfied: intel-cmplr-lib-rt in /usr/local/lib/python3.11/dist-packages (from mkl_umath->numpy->pycolmap) (2024.2.0)\r\n",
78
+ "Requirement already satisfied: intel-cmplr-lib-ur==2024.2.0 in /usr/local/lib/python3.11/dist-packages (from intel-openmp<2026,>=2024->mkl->numpy->pycolmap) (2024.2.0)\r\n",
79
+ "Downloading pycolmap-3.13.0-cp311-cp311-manylinux_2_28_x86_64.whl (20.3 MB)\r\n",
80
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m20.3/20.3 MB\u001b[0m \u001b[31m60.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\r\n",
81
+ "\u001b[?25hInstalling collected packages: pycolmap\r\n",
82
+ "Successfully installed pycolmap-3.13.0\r\n"
83
+ ]
84
+ }
85
+ ],
86
+ "source": [
87
+ "!pip install pycolmap"
88
+ ]
89
+ },
90
+ {
91
+ "cell_type": "code",
92
+ "execution_count": 2,
93
+ "id": "89e1896a",
94
+ "metadata": {
95
+ "execution": {
96
+ "iopub.execute_input": "2026-01-12T15:57:49.831872Z",
97
+ "iopub.status.busy": "2026-01-12T15:57:49.831155Z",
98
+ "iopub.status.idle": "2026-01-12T15:59:03.941461Z",
99
+ "shell.execute_reply": "2026-01-12T15:59:03.940181Z"
100
+ },
101
+ "papermill": {
102
+ "duration": 74.116273,
103
+ "end_time": "2026-01-12T15:59:03.943069",
104
+ "exception": false,
105
+ "start_time": "2026-01-12T15:57:49.826796",
106
+ "status": "completed"
107
+ },
108
+ "tags": []
109
+ },
110
+ "outputs": [
111
+ {
112
+ "name": "stdout",
113
+ "output_type": "stream",
114
+ "text": [
115
+ "✓ pycolmap is available\n",
116
+ "======================================================================\n",
117
+ "IMC2025 COLMAP PIPELINE v4\n",
118
+ "======================================================================\n",
119
+ "Loading competition data...\n",
120
+ "Testing with 73 entries from ['ETs', 'stairs']\n",
121
+ "Found 2 datasets\n",
122
+ "\n",
123
+ "======================================================================\n",
124
+ "DATASET: ETs\n",
125
+ "======================================================================\n",
126
+ "\n",
127
+ "============================================================\n",
128
+ "Processing: ETs\n",
129
+ "Images: 22\n",
130
+ "============================================================\n",
131
+ "Running COLMAP feature extraction...\n"
132
+ ]
133
+ },
134
+ {
135
+ "name": "stderr",
136
+ "output_type": "stream",
137
+ "text": [
138
+ "W20260112 15:57:52.003861 138282864959488 feature_extraction.cc:411] Your current options use the maximum number of threads on the machine to extract features. Extracting SIFT features on the CPU can consume a lot of RAM per thread for large images. Consider reducing the maximum image size and/or the first octave or manually limit the number of extraction threads. Ignore this warning, if your machine has sufficient memory for the current settings.\n",
139
+ "I20260112 15:57:52.004491 138281612211776 misc.cc:44] \n",
140
+ "==============================================================================\n",
141
+ "Feature extraction\n",
142
+ "==============================================================================\n",
143
+ "I20260112 15:57:52.004931 138281496864320 sift.cc:726] Creating SIFT CPU feature extractor\n",
144
+ "I20260112 15:57:52.004965 138281488471616 sift.cc:726] Creating SIFT CPU feature extractor\n",
145
+ "I20260112 15:57:52.005191 138281371039296 sift.cc:726] Creating SIFT CPU feature extractor\n",
146
+ "I20260112 15:57:52.005308 138281480078912 sift.cc:726] Creating SIFT CPU feature extractor\n",
147
+ "I20260112 15:57:52.013929 138281471686208 feature_extraction.cc:260] Processed file [1/23]\n",
148
+ "I20260112 15:57:52.013976 138281471686208 feature_extraction.cc:263] Name: LICENSE.txt\n",
149
+ "E20260112 15:57:52.013984 138281471686208 feature_extraction.cc:267] LICENSE.txt BITMAP_ERROR: Failed to read the image file format.\n",
150
+ "I20260112 15:57:52.979442 138281471686208 feature_extraction.cc:260] Processed file [2/23]\n",
151
+ "I20260112 15:57:52.979500 138281471686208 feature_extraction.cc:263] Name: another_et_another_et001.png\n",
152
+ "I20260112 15:57:52.979507 138281471686208 feature_extraction.cc:272] Dimensions: 360 x 640\n",
153
+ "I20260112 15:57:52.979513 138281471686208 feature_extraction.cc:275] Camera: #1 - SIMPLE_RADIAL\n",
154
+ "I20260112 15:57:52.979520 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
155
+ "I20260112 15:57:52.979534 138281471686208 feature_extraction.cc:282] Features: 2405 (SIFT)\n",
156
+ "I20260112 15:57:53.047437 138281471686208 feature_extraction.cc:260] Processed file [3/23]\n",
157
+ "I20260112 15:57:53.047493 138281471686208 feature_extraction.cc:263] Name: another_et_another_et002.png\n",
158
+ "I20260112 15:57:53.047502 138281471686208 feature_extraction.cc:272] Dimensions: 360 x 640\n",
159
+ "I20260112 15:57:53.047508 138281471686208 feature_extraction.cc:275] Camera: #2 - SIMPLE_RADIAL\n",
160
+ "I20260112 15:57:53.047525 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
161
+ "I20260112 15:57:53.047539 138281471686208 feature_extraction.cc:282] Features: 2389 (SIFT)\n",
162
+ "I20260112 15:57:53.214433 138281471686208 feature_extraction.cc:260] Processed file [4/23]\n",
163
+ "I20260112 15:57:53.214489 138281471686208 feature_extraction.cc:263] Name: another_et_another_et004.png\n",
164
+ "I20260112 15:57:53.214497 138281471686208 feature_extraction.cc:272] Dimensions: 360 x 640\n",
165
+ "I20260112 15:57:53.214504 138281471686208 feature_extraction.cc:275] Camera: #4 - SIMPLE_RADIAL\n",
166
+ "I20260112 15:57:53.214512 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
167
+ "I20260112 15:57:53.214526 138281471686208 feature_extraction.cc:282] Features: 2408 (SIFT)\n",
168
+ "I20260112 15:57:53.520430 138281471686208 feature_extraction.cc:260] Processed file [5/23]\n",
169
+ "I20260112 15:57:53.520492 138281471686208 feature_extraction.cc:263] Name: another_et_another_et003.png\n",
170
+ "I20260112 15:57:53.520500 138281471686208 feature_extraction.cc:272] Dimensions: 360 x 640\n",
171
+ "I20260112 15:57:53.520507 138281471686208 feature_extraction.cc:275] Camera: #3 - SIMPLE_RADIAL\n",
172
+ "I20260112 15:57:53.520521 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
173
+ "I20260112 15:57:53.520534 138281471686208 feature_extraction.cc:282] Features: 2463 (SIFT)\n",
174
+ "I20260112 15:57:53.833460 138281471686208 feature_extraction.cc:260] Processed file [6/23]\n",
175
+ "I20260112 15:57:53.833518 138281471686208 feature_extraction.cc:263] Name: another_et_another_et006.png\n",
176
+ "I20260112 15:57:53.833527 138281471686208 feature_extraction.cc:272] Dimensions: 360 x 640\n",
177
+ "I20260112 15:57:53.833533 138281471686208 feature_extraction.cc:275] Camera: #6 - SIMPLE_RADIAL\n",
178
+ "I20260112 15:57:53.833541 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
179
+ "I20260112 15:57:53.833556 138281471686208 feature_extraction.cc:282] Features: 2110 (SIFT)\n",
180
+ "I20260112 15:57:53.838434 138281471686208 feature_extraction.cc:260] Processed file [7/23]\n",
181
+ "I20260112 15:57:53.838860 138281471686208 feature_extraction.cc:263] Name: another_et_another_et005.png\n",
182
+ "I20260112 15:57:53.838994 138281471686208 feature_extraction.cc:272] Dimensions: 360 x 640\n",
183
+ "I20260112 15:57:53.839165 138281471686208 feature_extraction.cc:275] Camera: #5 - SIMPLE_RADIAL\n",
184
+ "I20260112 15:57:53.841423 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
185
+ "I20260112 15:57:53.841464 138281471686208 feature_extraction.cc:282] Features: 2353 (SIFT)\n",
186
+ "I20260112 15:57:54.146435 138281471686208 feature_extraction.cc:260] Processed file [8/23]\n",
187
+ "I20260112 15:57:54.146479 138281471686208 feature_extraction.cc:263] Name: another_et_another_et007.png\n",
188
+ "I20260112 15:57:54.146487 138281471686208 feature_extraction.cc:272] Dimensions: 360 x 640\n",
189
+ "I20260112 15:57:54.146494 138281471686208 feature_extraction.cc:275] Camera: #7 - SIMPLE_RADIAL\n",
190
+ "I20260112 15:57:54.146502 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
191
+ "I20260112 15:57:54.146515 138281471686208 feature_extraction.cc:282] Features: 2314 (SIFT)\n",
192
+ "I20260112 15:57:54.479347 138281471686208 feature_extraction.cc:260] Processed file [9/23]\n",
193
+ "I20260112 15:57:54.479948 138281471686208 feature_extraction.cc:263] Name: another_et_another_et010.png\n",
194
+ "I20260112 15:57:54.480042 138281471686208 feature_extraction.cc:272] Dimensions: 360 x 640\n",
195
+ "I20260112 15:57:54.480095 138281471686208 feature_extraction.cc:275] Camera: #10 - SIMPLE_RADIAL\n",
196
+ "I20260112 15:57:54.480153 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
197
+ "I20260112 15:57:54.480209 138281471686208 feature_extraction.cc:282] Features: 1260 (SIFT)\n",
198
+ "I20260112 15:57:54.547603 138281471686208 feature_extraction.cc:260] Processed file [10/23]\n",
199
+ "I20260112 15:57:54.547648 138281471686208 feature_extraction.cc:263] Name: another_et_another_et008.png\n",
200
+ "I20260112 15:57:54.547655 138281471686208 feature_extraction.cc:272] Dimensions: 360 x 640\n",
201
+ "I20260112 15:57:54.547661 138281471686208 feature_extraction.cc:275] Camera: #8 - SIMPLE_RADIAL\n",
202
+ "I20260112 15:57:54.547668 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
203
+ "I20260112 15:57:54.547681 138281471686208 feature_extraction.cc:282] Features: 2123 (SIFT)\n",
204
+ "I20260112 15:57:54.641433 138281471686208 feature_extraction.cc:260] Processed file [11/23]\n",
205
+ "I20260112 15:57:54.641485 138281471686208 feature_extraction.cc:263] Name: another_et_another_et009.png\n",
206
+ "I20260112 15:57:54.641493 138281471686208 feature_extraction.cc:272] Dimensions: 360 x 640\n",
207
+ "I20260112 15:57:54.641499 138281471686208 feature_extraction.cc:275] Camera: #9 - SIMPLE_RADIAL\n",
208
+ "I20260112 15:57:54.641508 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
209
+ "I20260112 15:57:54.641522 138281471686208 feature_extraction.cc:282] Features: 1589 (SIFT)\n",
210
+ "I20260112 15:57:55.267649 138281471686208 feature_extraction.cc:260] Processed file [12/23]\n",
211
+ "I20260112 15:57:55.267715 138281471686208 feature_extraction.cc:263] Name: et_et000.png\n",
212
+ "I20260112 15:57:55.267723 138281471686208 feature_extraction.cc:272] Dimensions: 480 x 640\n",
213
+ "I20260112 15:57:55.267730 138281471686208 feature_extraction.cc:275] Camera: #11 - SIMPLE_RADIAL\n",
214
+ "I20260112 15:57:55.267738 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
215
+ "I20260112 15:57:55.267753 138281471686208 feature_extraction.cc:282] Features: 2398 (SIFT)\n",
216
+ "I20260112 15:57:55.408434 138281471686208 feature_extraction.cc:260] Processed file [13/23]\n",
217
+ "I20260112 15:57:55.408497 138281471686208 feature_extraction.cc:263] Name: et_et002.png\n",
218
+ "I20260112 15:57:55.408505 138281471686208 feature_extraction.cc:272] Dimensions: 480 x 640\n",
219
+ "I20260112 15:57:55.408511 138281471686208 feature_extraction.cc:275] Camera: #13 - SIMPLE_RADIAL\n",
220
+ "I20260112 15:57:55.408520 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
221
+ "I20260112 15:57:55.408533 138281471686208 feature_extraction.cc:282] Features: 1906 (SIFT)\n",
222
+ "I20260112 15:57:55.918441 138281471686208 feature_extraction.cc:260] Processed file [14/23]\n",
223
+ "I20260112 15:57:55.918499 138281471686208 feature_extraction.cc:263] Name: et_et001.png\n",
224
+ "I20260112 15:57:55.918512 138281471686208 feature_extraction.cc:272] Dimensions: 480 x 640\n",
225
+ "I20260112 15:57:55.918518 138281471686208 feature_extraction.cc:275] Camera: #12 - SIMPLE_RADIAL\n",
226
+ "I20260112 15:57:55.918526 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
227
+ "I20260112 15:57:55.918540 138281471686208 feature_extraction.cc:282] Features: 2187 (SIFT)\n",
228
+ "I20260112 15:57:55.937368 138281471686208 feature_extraction.cc:260] Processed file [15/23]\n",
229
+ "I20260112 15:57:55.937731 138281471686208 feature_extraction.cc:263] Name: et_et003.png\n",
230
+ "I20260112 15:57:55.937785 138281471686208 feature_extraction.cc:272] Dimensions: 480 x 640\n",
231
+ "I20260112 15:57:55.937818 138281471686208 feature_extraction.cc:275] Camera: #14 - SIMPLE_RADIAL\n",
232
+ "I20260112 15:57:55.937888 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
233
+ "I20260112 15:57:55.937923 138281471686208 feature_extraction.cc:282] Features: 2578 (SIFT)\n",
234
+ "I20260112 15:57:56.309455 138281471686208 feature_extraction.cc:260] Processed file [16/23]\n",
235
+ "I20260112 15:57:56.309512 138281471686208 feature_extraction.cc:263] Name: et_et005.png\n",
236
+ "I20260112 15:57:56.309520 138281471686208 feature_extraction.cc:272] Dimensions: 480 x 640\n",
237
+ "I20260112 15:57:56.309528 138281471686208 feature_extraction.cc:275] Camera: #16 - SIMPLE_RADIAL\n",
238
+ "I20260112 15:57:56.309537 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
239
+ "I20260112 15:57:56.309551 138281471686208 feature_extraction.cc:282] Features: 2080 (SIFT)\n",
240
+ "I20260112 15:57:56.521436 138281471686208 feature_extraction.cc:260] Processed file [17/23]\n",
241
+ "I20260112 15:57:56.521491 138281471686208 feature_extraction.cc:263] Name: et_et004.png\n",
242
+ "I20260112 15:57:56.521500 138281471686208 feature_extraction.cc:272] Dimensions: 480 x 640\n",
243
+ "I20260112 15:57:56.521506 138281471686208 feature_extraction.cc:275] Camera: #15 - SIMPLE_RADIAL\n",
244
+ "I20260112 15:57:56.521514 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
245
+ "I20260112 15:57:56.521528 138281471686208 feature_extraction.cc:282] Features: 3250 (SIFT)\n",
246
+ "I20260112 15:57:56.836042 138281471686208 feature_extraction.cc:260] Processed file [18/23]\n",
247
+ "I20260112 15:57:56.836104 138281471686208 feature_extraction.cc:263] Name: et_et007.png\n",
248
+ "I20260112 15:57:56.836112 138281471686208 feature_extraction.cc:272] Dimensions: 480 x 640\n",
249
+ "I20260112 15:57:56.836118 138281471686208 feature_extraction.cc:275] Camera: #18 - SIMPLE_RADIAL\n",
250
+ "I20260112 15:57:56.836126 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
251
+ "I20260112 15:57:56.836161 138281471686208 feature_extraction.cc:282] Features: 1798 (SIFT)\n",
252
+ "I20260112 15:57:56.919432 138281471686208 feature_extraction.cc:260] Processed file [19/23]\n",
253
+ "I20260112 15:57:56.919488 138281471686208 feature_extraction.cc:263] Name: et_et006.png\n",
254
+ "I20260112 15:57:56.919496 138281471686208 feature_extraction.cc:272] Dimensions: 480 x 640\n",
255
+ "I20260112 15:57:56.919502 138281471686208 feature_extraction.cc:275] Camera: #17 - SIMPLE_RADIAL\n",
256
+ "I20260112 15:57:56.919510 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
257
+ "I20260112 15:57:56.919523 138281471686208 feature_extraction.cc:282] Features: 1954 (SIFT)\n",
258
+ "I20260112 15:57:57.177437 138281471686208 feature_extraction.cc:260] Processed file [20/23]\n",
259
+ "I20260112 15:57:57.177491 138281471686208 feature_extraction.cc:263] Name: outliers_out_et001.png\n",
260
+ "I20260112 15:57:57.177499 138281471686208 feature_extraction.cc:272] Dimensions: 262 x 450\n",
261
+ "I20260112 15:57:57.177505 138281471686208 feature_extraction.cc:275] Camera: #20 - SIMPLE_RADIAL\n",
262
+ "I20260112 15:57:57.177512 138281471686208 feature_extraction.cc:278] Focal Length: 540.00px\n",
263
+ "I20260112 15:57:57.177565 138281471686208 feature_extraction.cc:282] Features: 695 (SIFT)\n",
264
+ "I20260112 15:57:57.235442 138281471686208 feature_extraction.cc:260] Processed file [21/23]\n",
265
+ "I20260112 15:57:57.235499 138281471686208 feature_extraction.cc:263] Name: et_et008.png\n",
266
+ "I20260112 15:57:57.235507 138281471686208 feature_extraction.cc:272] Dimensions: 480 x 640\n",
267
+ "I20260112 15:57:57.235513 138281471686208 feature_extraction.cc:275] Camera: #19 - SIMPLE_RADIAL\n",
268
+ "I20260112 15:57:57.235520 138281471686208 feature_extraction.cc:278] Focal Length: 768.00px\n",
269
+ "I20260112 15:57:57.235534 138281471686208 feature_extraction.cc:282] Features: 2138 (SIFT)\n",
270
+ "I20260112 15:57:57.245523 138281471686208 feature_extraction.cc:260] Processed file [22/23]\n",
271
+ "I20260112 15:57:57.245606 138281471686208 feature_extraction.cc:263] Name: outliers_out_et002.png\n",
272
+ "I20260112 15:57:57.245620 138281471686208 feature_extraction.cc:272] Dimensions: 300 x 300\n",
273
+ "I20260112 15:57:57.245627 138281471686208 feature_extraction.cc:275] Camera: #21 - SIMPLE_RADIAL\n",
274
+ "I20260112 15:57:57.245635 138281471686208 feature_extraction.cc:278] Focal Length: 360.00px\n",
275
+ "I20260112 15:57:57.245648 138281471686208 feature_extraction.cc:282] Features: 347 (SIFT)\n",
276
+ "I20260112 15:57:57.518653 138281471686208 feature_extraction.cc:260] Processed file [23/23]\n",
277
+ "I20260112 15:57:57.518718 138281471686208 feature_extraction.cc:263] Name: outliers_out_et003.png\n",
278
+ "I20260112 15:57:57.518726 138281471686208 feature_extraction.cc:272] Dimensions: 344 x 500\n",
279
+ "I20260112 15:57:57.518733 138281471686208 feature_extraction.cc:275] Camera: #22 - SIMPLE_RADIAL\n",
280
+ "I20260112 15:57:57.518740 138281471686208 feature_extraction.cc:278] Focal Length: 600.00px\n",
281
+ "I20260112 15:57:57.518755 138281471686208 feature_extraction.cc:282] Features: 1844 (SIFT)\n",
282
+ "I20260112 15:57:57.520004 138281612211776 timer.cc:90] Elapsed time: 0.092 [minutes]\n",
283
+ "I20260112 15:57:57.528989 138281612211776 misc.cc:44] \n",
284
+ "==============================================================================\n",
285
+ "Feature matching & geometric verification\n",
286
+ "==============================================================================\n",
287
+ "I20260112 15:57:57.529262 138281471686208 sift.cc:1452] Creating SIFT CPU feature matcher\n",
288
+ "I20260112 15:57:57.529294 138281603819072 sift.cc:1452] Creating SIFT CPU feature matcher\n",
289
+ "I20260112 15:57:57.529582 138281480078912 sift.cc:1452] Creating SIFT CPU feature matcher\n",
290
+ "I20260112 15:57:57.529643 138281371039296 sift.cc:1452] Creating SIFT CPU feature matcher\n",
291
+ "I20260112 15:57:57.529857 138281612211776 pairing.cc:180] Generating exhaustive image pairs...\n",
292
+ "I20260112 15:57:57.529876 138281612211776 pairing.cc:213] Processing block [1/1, 1/1]\n"
293
+ ]
294
+ },
295
+ {
296
+ "name": "stdout",
297
+ "output_type": "stream",
298
+ "text": [
299
+ " Extracted features from 22 images\n",
300
+ "Running COLMAP feature matching...\n"
301
+ ]
302
+ },
303
+ {
304
+ "name": "stderr",
305
+ "output_type": "stream",
306
+ "text": [
307
+ "I20260112 15:58:00.946895 138281612211776 feature_matching.cc:117] in 3.417s\n",
308
+ "I20260112 15:58:00.946945 138281612211776 timer.cc:90] Elapsed time: 0.057 [minutes]\n",
309
+ "Traceback (most recent call last):\n",
310
+ " File \"/tmp/ipykernel_13/3332504298.py\", line 204, in run_colmap_mapper\n",
311
+ " maps = pycolmap.incremental_mapping(\n",
312
+ " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n",
313
+ "TypeError: incremental_mapping(): incompatible function arguments. The following argument types are supported:\n",
314
+ " 1. (database_path: str, image_path: str, output_path: str, options: pycolmap._core.IncrementalPipelineOptions = IncrementalPipelineOptions(), input_path: str = '', initial_image_pair_callback: collections.abc.Callable[[], None] = None, next_image_callback: collections.abc.Callable[[], None] = None) -> dict[int, pycolmap._core.Reconstruction]\n",
315
+ "\n",
316
+ "Invoked with: kwargs: database_path='/kaggle/working/result/features/ETs/database.db', image_path='/kaggle/input/image-matching-challenge-2025/test/ETs', output_path='/kaggle/working/result/reconstructions/ETs', options=IncrementalMapperOptions(init_min_num_inliers=100, init_max_error=4.0, init_max_forward_motion=0.95, init_min_tri_angle=16.0, init_max_reg_trials=2, abs_pose_max_error=12.0, abs_pose_min_num_inliers=30, abs_pose_min_inlier_ratio=0.25, abs_pose_refine_focal_length=True, abs_pose_refine_extra_params=True, ba_local_num_images=6, ba_local_min_tri_angle=6.0, ba_global_ignore_redundant_points3D=False, ba_global_prune_points_min_coverage_gain=0.05, min_focal_length_ratio=0.1, max_focal_length_ratio=10.0, max_extra_param=1.0, filter_max_reproj_error=4.0, filter_min_tri_angle=1.5, max_reg_trials=3, fix_existing_frames=False, constant_rigs=set(), constant_cameras=set(), num_threads=-1, random_seed=-1, image_selection_method=ImageSelectionMethod.MIN_UNCERTAINTY)\n"
317
+ ]
318
+ },
319
+ {
320
+ "name": "stdout",
321
+ "output_type": "stream",
322
+ "text": [
323
+ " Feature matching completed\n",
324
+ "Running COLMAP mapper...\n",
325
+ " Mapper failed: incremental_mapping(): incompatible function arguments. The following argument types are supported:\n",
326
+ " 1. (database_path: str, image_path: str, output_path: str, options: pycolmap._core.IncrementalPipelineOptions = IncrementalPipelineOptions(), input_path: str = '', initial_image_pair_callback: collections.abc.Callable[[], None] = None, next_image_callback: collections.abc.Callable[[], None] = None) -> dict[int, pycolmap._core.Reconstruction]\n",
327
+ "\n",
328
+ "Invoked with: kwargs: database_path='/kaggle/working/result/features/ETs/database.db', image_path='/kaggle/input/image-matching-challenge-2025/test/ETs', output_path='/kaggle/working/result/reconstructions/ETs', options=IncrementalMapperOptions(init_min_num_inliers=100, init_max_error=4.0, init_max_forward_motion=0.95, init_min_tri_angle=16.0, init_max_reg_trials=2, abs_pose_max_error=12.0, abs_pose_min_num_inliers=30, abs_pose_min_inlier_ratio=0.25, abs_pose_refine_focal_length=True, abs_pose_refine_extra_params=True, ba_local_num_images=6, ba_local_min_tri_angle=6.0, ba_global_ignore_redundant_points3D=False, ba_global_prune_points_min_coverage_gain=0.05, min_focal_length_ratio=0.1, max_focal_length_ratio=10.0, max_extra_param=1.0, filter_max_reproj_error=4.0, filter_min_tri_angle=1.5, max_reg_trials=3, fix_existing_frames=False, constant_rigs=set(), constant_cameras=set(), num_threads=-1, random_seed=-1, image_selection_method=ImageSelectionMethod.MIN_UNCERTAINTY)\n",
329
+ "✅ Completed ETs\n",
330
+ "\n",
331
+ "======================================================================\n",
332
+ "DATASET: stairs\n",
333
+ "======================================================================\n",
334
+ "\n",
335
+ "============================================================\n",
336
+ "Processing: stairs\n",
337
+ "Images: 51\n",
338
+ "============================================================\n",
339
+ "Running COLMAP feature extraction...\n"
340
+ ]
341
+ },
342
+ {
343
+ "name": "stderr",
344
+ "output_type": "stream",
345
+ "text": [
346
+ "W20260112 15:58:01.270529 138282864959488 feature_extraction.cc:411] Your current options use the maximum number of threads on the machine to extract features. Extracting SIFT features on the CPU can consume a lot of RAM per thread for large images. Consider reducing the maximum image size and/or the first octave or manually limit the number of extraction threads. Ignore this warning, if your machine has sufficient memory for the current settings.\n",
347
+ "I20260112 15:58:01.270747 138281496864320 misc.cc:44] \n",
348
+ "==============================================================================\n",
349
+ "Feature extraction\n",
350
+ "==============================================================================\n",
351
+ "I20260112 15:58:01.271180 138281603819072 sift.cc:726] Creating SIFT CPU feature extractor\n",
352
+ "I20260112 15:58:01.271231 138281488471616 sift.cc:726] Creating SIFT CPU feature extractor\n",
353
+ "I20260112 15:58:01.271249 138281480078912 sift.cc:726] Creating SIFT CPU feature extractor\n",
354
+ "I20260112 15:58:01.271315 138281471686208 sift.cc:726] Creating SIFT CPU feature extractor\n",
355
+ "I20260112 15:58:01.277715 138281463293504 feature_extraction.cc:260] Processed file [1/52]\n",
356
+ "I20260112 15:58:01.277797 138281463293504 feature_extraction.cc:263] Name: LICENSE.txt\n",
357
+ "E20260112 15:58:01.277804 138281463293504 feature_extraction.cc:267] LICENSE.txt BITMAP_ERROR: Failed to read the image file format.\n",
358
+ "I20260112 15:58:05.434373 138281463293504 feature_extraction.cc:260] Processed file [2/52]\n",
359
+ "I20260112 15:58:05.434442 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453601885.png\n",
360
+ "I20260112 15:58:05.434451 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
361
+ "I20260112 15:58:05.434456 138281463293504 feature_extraction.cc:275] Camera: #2 - SIMPLE_RADIAL\n",
362
+ "I20260112 15:58:05.434464 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
363
+ "I20260112 15:58:05.434478 138281463293504 feature_extraction.cc:282] Features: 1208 (SIFT)\n",
364
+ "I20260112 15:58:05.783566 138281463293504 feature_extraction.cc:260] Processed file [3/52]\n",
365
+ "I20260112 15:58:05.783620 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453576271.png\n",
366
+ "I20260112 15:58:05.783628 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
367
+ "I20260112 15:58:05.783633 138281463293504 feature_extraction.cc:275] Camera: #1 - SIMPLE_RADIAL\n",
368
+ "I20260112 15:58:05.783641 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
369
+ "I20260112 15:58:05.783654 138281463293504 feature_extraction.cc:282] Features: 1641 (SIFT)\n",
370
+ "I20260112 15:58:05.905436 138281463293504 feature_extraction.cc:260] Processed file [4/52]\n",
371
+ "I20260112 15:58:05.905489 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453606287.png\n",
372
+ "I20260112 15:58:05.905497 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
373
+ "I20260112 15:58:05.905504 138281463293504 feature_extraction.cc:275] Camera: #3 - SIMPLE_RADIAL\n",
374
+ "I20260112 15:58:05.905512 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
375
+ "I20260112 15:58:05.905527 138281463293504 feature_extraction.cc:282] Features: 1358 (SIFT)\n",
376
+ "I20260112 15:58:06.018969 138281463293504 feature_extraction.cc:260] Processed file [5/52]\n",
377
+ "I20260112 15:58:06.019542 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453612890.png\n",
378
+ "I20260112 15:58:06.019591 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
379
+ "I20260112 15:58:06.019623 138281463293504 feature_extraction.cc:275] Camera: #4 - SIMPLE_RADIAL\n",
380
+ "I20260112 15:58:06.019657 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
381
+ "I20260112 15:58:06.019686 138281463293504 feature_extraction.cc:282] Features: 713 (SIFT)\n",
382
+ "I20260112 15:58:08.900585 138281463293504 feature_extraction.cc:260] Processed file [6/52]\n",
383
+ "I20260112 15:58:08.901052 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453616892.png\n",
384
+ "I20260112 15:58:08.901100 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
385
+ "I20260112 15:58:08.901125 138281463293504 feature_extraction.cc:275] Camera: #5 - SIMPLE_RADIAL\n",
386
+ "I20260112 15:58:08.901148 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
387
+ "I20260112 15:58:08.901176 138281463293504 feature_extraction.cc:282] Features: 1379 (SIFT)\n",
388
+ "I20260112 15:58:09.205434 138281463293504 feature_extraction.cc:260] Processed file [7/52]\n",
389
+ "I20260112 15:58:09.205919 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453620694.png\n",
390
+ "I20260112 15:58:09.206175 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
391
+ "I20260112 15:58:09.206278 138281463293504 feature_extraction.cc:275] Camera: #6 - SIMPLE_RADIAL\n",
392
+ "I20260112 15:58:09.206354 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
393
+ "I20260112 15:58:09.206437 138281463293504 feature_extraction.cc:282] Features: 900 (SIFT)\n",
394
+ "I20260112 15:58:09.444447 138281463293504 feature_extraction.cc:260] Processed file [8/52]\n",
395
+ "I20260112 15:58:09.444513 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453626698.png\n",
396
+ "I20260112 15:58:09.444603 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
397
+ "I20260112 15:58:09.444616 138281463293504 feature_extraction.cc:275] Camera: #7 - SIMPLE_RADIAL\n",
398
+ "I20260112 15:58:09.444625 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
399
+ "I20260112 15:58:09.444643 138281463293504 feature_extraction.cc:282] Features: 498 (SIFT)\n",
400
+ "I20260112 15:58:09.705443 138281463293504 feature_extraction.cc:260] Processed file [9/52]\n",
401
+ "I20260112 15:58:09.705490 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453643106.png\n",
402
+ "I20260112 15:58:09.705497 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
403
+ "I20260112 15:58:09.705503 138281463293504 feature_extraction.cc:275] Camera: #8 - SIMPLE_RADIAL\n",
404
+ "I20260112 15:58:09.705510 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
405
+ "I20260112 15:58:09.705522 138281463293504 feature_extraction.cc:282] Features: 1762 (SIFT)\n",
406
+ "I20260112 15:58:12.465441 138281463293504 feature_extraction.cc:260] Processed file [10/52]\n",
407
+ "I20260112 15:58:12.465492 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453651110.png\n",
408
+ "I20260112 15:58:12.465500 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
409
+ "I20260112 15:58:12.465506 138281463293504 feature_extraction.cc:275] Camera: #9 - SIMPLE_RADIAL\n",
410
+ "I20260112 15:58:12.465514 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
411
+ "I20260112 15:58:12.465528 138281463293504 feature_extraction.cc:282] Features: 966 (SIFT)\n",
412
+ "I20260112 15:58:12.660816 138281463293504 feature_extraction.cc:260] Processed file [11/52]\n",
413
+ "I20260112 15:58:12.661333 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453667117.png\n",
414
+ "I20260112 15:58:12.661585 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
415
+ "I20260112 15:58:12.661843 138281463293504 feature_extraction.cc:275] Camera: #12 - SIMPLE_RADIAL\n",
416
+ "I20260112 15:58:12.664423 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
417
+ "I20260112 15:58:12.664459 138281463293504 feature_extraction.cc:282] Features: 64 (SIFT)\n",
418
+ "I20260112 15:58:12.824112 138281463293504 feature_extraction.cc:260] Processed file [12/52]\n",
419
+ "I20260112 15:58:12.824177 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453659313.png\n",
420
+ "I20260112 15:58:12.824193 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
421
+ "I20260112 15:58:12.824205 138281463293504 feature_extraction.cc:275] Camera: #10 - SIMPLE_RADIAL\n",
422
+ "I20260112 15:58:12.824218 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
423
+ "I20260112 15:58:12.824240 138281463293504 feature_extraction.cc:282] Features: 1446 (SIFT)\n",
424
+ "I20260112 15:58:13.114440 138281463293504 feature_extraction.cc:260] Processed file [13/52]\n",
425
+ "I20260112 15:58:13.114502 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453663515.png\n",
426
+ "I20260112 15:58:13.114510 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
427
+ "I20260112 15:58:13.114516 138281463293504 feature_extraction.cc:275] Camera: #11 - SIMPLE_RADIAL\n",
428
+ "I20260112 15:58:13.114523 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
429
+ "I20260112 15:58:13.114537 138281463293504 feature_extraction.cc:282] Features: 365 (SIFT)\n",
430
+ "I20260112 15:58:15.893249 138281463293504 feature_extraction.cc:260] Processed file [14/52]\n",
431
+ "I20260112 15:58:15.893305 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453675921.png\n",
432
+ "I20260112 15:58:15.893313 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
433
+ "I20260112 15:58:15.893319 138281463293504 feature_extraction.cc:275] Camera: #14 - SIMPLE_RADIAL\n",
434
+ "I20260112 15:58:15.893337 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
435
+ "I20260112 15:58:15.893351 138281463293504 feature_extraction.cc:282] Features: 323 (SIFT)\n",
436
+ "I20260112 15:58:16.209429 138281463293504 feature_extraction.cc:260] Processed file [15/52]\n",
437
+ "I20260112 15:58:16.209478 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453668718.png\n",
438
+ "I20260112 15:58:16.209486 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
439
+ "I20260112 15:58:16.209493 138281463293504 feature_extraction.cc:275] Camera: #13 - SIMPLE_RADIAL\n",
440
+ "I20260112 15:58:16.209501 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
441
+ "I20260112 15:58:16.209514 138281463293504 feature_extraction.cc:282] Features: 427 (SIFT)\n",
442
+ "I20260112 15:58:16.405007 138281463293504 feature_extraction.cc:260] Processed file [16/52]\n",
443
+ "I20260112 15:58:16.405068 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453683725.png\n",
444
+ "I20260112 15:58:16.405076 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
445
+ "I20260112 15:58:16.405082 138281463293504 feature_extraction.cc:275] Camera: #16 - SIMPLE_RADIAL\n",
446
+ "I20260112 15:58:16.405090 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
447
+ "I20260112 15:58:16.405104 138281463293504 feature_extraction.cc:282] Features: 1277 (SIFT)\n",
448
+ "I20260112 15:58:16.758368 138281463293504 feature_extraction.cc:260] Processed file [17/52]\n",
449
+ "I20260112 15:58:16.758446 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453678922.png\n",
450
+ "I20260112 15:58:16.758456 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
451
+ "I20260112 15:58:16.758465 138281463293504 feature_extraction.cc:275] Camera: #15 - SIMPLE_RADIAL\n",
452
+ "I20260112 15:58:16.758475 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
453
+ "I20260112 15:58:16.758490 138281463293504 feature_extraction.cc:282] Features: 1564 (SIFT)\n",
454
+ "I20260112 15:58:19.839427 138281463293504 feature_extraction.cc:260] Processed file [18/52]\n",
455
+ "I20260112 15:58:19.839485 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453689727.png\n",
456
+ "I20260112 15:58:19.839494 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
457
+ "I20260112 15:58:19.839500 138281463293504 feature_extraction.cc:275] Camera: #17 - SIMPLE_RADIAL\n",
458
+ "I20260112 15:58:19.839509 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
459
+ "I20260112 15:58:19.839523 138281463293504 feature_extraction.cc:282] Features: 3510 (SIFT)\n",
460
+ "I20260112 15:58:19.963429 138281463293504 feature_extraction.cc:260] Processed file [19/52]\n",
461
+ "I20260112 15:58:19.963482 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453693529.png\n",
462
+ "I20260112 15:58:19.963490 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
463
+ "I20260112 15:58:19.963497 138281463293504 feature_extraction.cc:275] Camera: #18 - SIMPLE_RADIAL\n",
464
+ "I20260112 15:58:19.963505 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
465
+ "I20260112 15:58:19.963519 138281463293504 feature_extraction.cc:282] Features: 2192 (SIFT)\n",
466
+ "I20260112 15:58:20.441473 138281463293504 feature_extraction.cc:260] Processed file [20/52]\n",
467
+ "I20260112 15:58:20.441532 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453697531.png\n",
468
+ "I20260112 15:58:20.441540 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
469
+ "I20260112 15:58:20.441546 138281463293504 feature_extraction.cc:275] Camera: #19 - SIMPLE_RADIAL\n",
470
+ "I20260112 15:58:20.441554 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
471
+ "I20260112 15:58:20.441568 138281463293504 feature_extraction.cc:282] Features: 3870 (SIFT)\n",
472
+ "I20260112 15:58:20.471717 138281463293504 feature_extraction.cc:260] Processed file [21/52]\n",
473
+ "I20260112 15:58:20.471772 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453704934.png\n",
474
+ "I20260112 15:58:20.471781 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
475
+ "I20260112 15:58:20.471787 138281463293504 feature_extraction.cc:275] Camera: #20 - SIMPLE_RADIAL\n",
476
+ "I20260112 15:58:20.471849 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
477
+ "I20260112 15:58:20.471866 138281463293504 feature_extraction.cc:282] Features: 1476 (SIFT)\n",
478
+ "I20260112 15:58:23.586104 138281463293504 feature_extraction.cc:260] Processed file [22/52]\n",
479
+ "I20260112 15:58:23.586209 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453901046.png\n",
480
+ "I20260112 15:58:23.586221 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
481
+ "I20260112 15:58:23.586227 138281463293504 feature_extraction.cc:275] Camera: #21 - SIMPLE_RADIAL\n",
482
+ "I20260112 15:58:23.586235 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
483
+ "I20260112 15:58:23.586249 138281463293504 feature_extraction.cc:282] Features: 1739 (SIFT)\n",
484
+ "I20260112 15:58:23.731900 138281463293504 feature_extraction.cc:260] Processed file [23/52]\n",
485
+ "I20260112 15:58:23.731958 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453930259.png\n",
486
+ "I20260112 15:58:23.731966 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
487
+ "I20260112 15:58:23.731972 138281463293504 feature_extraction.cc:275] Camera: #23 - SIMPLE_RADIAL\n",
488
+ "I20260112 15:58:23.731980 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
489
+ "I20260112 15:58:23.731994 138281463293504 feature_extraction.cc:282] Features: 584 (SIFT)\n",
490
+ "I20260112 15:58:23.761413 138281463293504 feature_extraction.cc:260] Processed file [24/52]\n",
491
+ "I20260112 15:58:23.761957 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453912451.png\n",
492
+ "I20260112 15:58:23.762264 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
493
+ "I20260112 15:58:23.762498 138281463293504 feature_extraction.cc:275] Camera: #22 - SIMPLE_RADIAL\n",
494
+ "I20260112 15:58:23.762762 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
495
+ "I20260112 15:58:23.762948 138281463293504 feature_extraction.cc:282] Features: 2032 (SIFT)\n",
496
+ "I20260112 15:58:24.409442 138281463293504 feature_extraction.cc:260] Processed file [25/52]\n",
497
+ "I20260112 15:58:24.409499 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453947066.png\n",
498
+ "I20260112 15:58:24.409508 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
499
+ "I20260112 15:58:24.409516 138281463293504 feature_extraction.cc:275] Camera: #24 - SIMPLE_RADIAL\n",
500
+ "I20260112 15:58:24.409526 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
501
+ "I20260112 15:58:24.409549 138281463293504 feature_extraction.cc:282] Features: 832 (SIFT)\n",
502
+ "I20260112 15:58:27.025574 138281463293504 feature_extraction.cc:260] Processed file [26/52]\n",
503
+ "I20260112 15:58:27.025645 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453985484.png\n",
504
+ "I20260112 15:58:27.025657 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
505
+ "I20260112 15:58:27.025667 138281463293504 feature_extraction.cc:275] Camera: #27 - SIMPLE_RADIAL\n",
506
+ "I20260112 15:58:27.025679 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
507
+ "I20260112 15:58:27.025698 138281463293504 feature_extraction.cc:282] Features: 579 (SIFT)\n",
508
+ "I20260112 15:58:27.339439 138281463293504 feature_extraction.cc:260] Processed file [27/52]\n",
509
+ "I20260112 15:58:27.339492 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453963274.png\n",
510
+ "I20260112 15:58:27.339500 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
511
+ "I20260112 15:58:27.339506 138281463293504 feature_extraction.cc:275] Camera: #26 - SIMPLE_RADIAL\n",
512
+ "I20260112 15:58:27.339514 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
513
+ "I20260112 15:58:27.339528 138281463293504 feature_extraction.cc:282] Features: 800 (SIFT)\n",
514
+ "I20260112 15:58:27.652157 138281463293504 feature_extraction.cc:260] Processed file [28/52]\n",
515
+ "I20260112 15:58:27.652216 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453955270.png\n",
516
+ "I20260112 15:58:27.652223 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
517
+ "I20260112 15:58:27.652229 138281463293504 feature_extraction.cc:275] Camera: #25 - SIMPLE_RADIAL\n",
518
+ "I20260112 15:58:27.652236 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
519
+ "I20260112 15:58:27.652294 138281463293504 feature_extraction.cc:282] Features: 2733 (SIFT)\n",
520
+ "I20260112 15:58:27.884433 138281463293504 feature_extraction.cc:260] Processed file [29/52]\n",
521
+ "I20260112 15:58:27.884492 138281463293504 feature_extraction.cc:263] Name: stairs_split_1_1710453990286.png\n",
522
+ "I20260112 15:58:27.884500 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
523
+ "I20260112 15:58:27.884506 138281463293504 feature_extraction.cc:275] Camera: #28 - SIMPLE_RADIAL\n",
524
+ "I20260112 15:58:27.884513 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
525
+ "I20260112 15:58:27.884525 138281463293504 feature_extraction.cc:282] Features: 1091 (SIFT)\n",
526
+ "I20260112 15:58:30.844331 138281463293504 feature_extraction.cc:260] Processed file [30/52]\n",
527
+ "I20260112 15:58:30.844388 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453725143.png\n",
528
+ "I20260112 15:58:30.844411 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
529
+ "I20260112 15:58:30.844419 138281463293504 feature_extraction.cc:275] Camera: #30 - SIMPLE_RADIAL\n",
530
+ "I20260112 15:58:30.844427 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
531
+ "I20260112 15:58:30.844441 138281463293504 feature_extraction.cc:282] Features: 1914 (SIFT)\n",
532
+ "I20260112 15:58:30.978173 138281463293504 feature_extraction.cc:260] Processed file [31/52]\n",
533
+ "I20260112 15:58:30.978822 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453720741.png\n",
534
+ "I20260112 15:58:30.979073 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
535
+ "I20260112 15:58:30.979283 138281463293504 feature_extraction.cc:275] Camera: #29 - SIMPLE_RADIAL\n",
536
+ "I20260112 15:58:30.979573 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
537
+ "I20260112 15:58:30.979785 138281463293504 feature_extraction.cc:282] Features: 3008 (SIFT)\n",
538
+ "I20260112 15:58:31.539703 138281463293504 feature_extraction.cc:260] Processed file [32/52]\n",
539
+ "I20260112 15:58:31.540319 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453733751.png\n",
540
+ "I20260112 15:58:31.540761 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
541
+ "I20260112 15:58:31.540952 138281463293504 feature_extraction.cc:275] Camera: #32 - SIMPLE_RADIAL\n",
542
+ "I20260112 15:58:31.541179 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
543
+ "I20260112 15:58:31.541385 138281463293504 feature_extraction.cc:282] Features: 1750 (SIFT)\n",
544
+ "I20260112 15:58:31.799216 138281463293504 feature_extraction.cc:260] Processed file [33/52]\n",
545
+ "I20260112 15:58:31.799263 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453728949.png\n",
546
+ "I20260112 15:58:31.799269 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
547
+ "I20260112 15:58:31.799275 138281463293504 feature_extraction.cc:275] Camera: #31 - SIMPLE_RADIAL\n",
548
+ "I20260112 15:58:31.799282 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
549
+ "I20260112 15:58:31.799295 138281463293504 feature_extraction.cc:282] Features: 1918 (SIFT)\n",
550
+ "I20260112 15:58:34.612430 138281463293504 feature_extraction.cc:260] Processed file [34/52]\n",
551
+ "I20260112 15:58:34.612482 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453739354.png\n",
552
+ "I20260112 15:58:34.612490 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
553
+ "I20260112 15:58:34.612497 138281463293504 feature_extraction.cc:275] Camera: #34 - SIMPLE_RADIAL\n",
554
+ "I20260112 15:58:34.612505 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
555
+ "I20260112 15:58:34.612519 138281463293504 feature_extraction.cc:282] Features: 1096 (SIFT)\n",
556
+ "I20260112 15:58:34.889432 138281463293504 feature_extraction.cc:260] Processed file [35/52]\n",
557
+ "I20260112 15:58:34.889481 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453736752.png\n",
558
+ "I20260112 15:58:34.889489 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
559
+ "I20260112 15:58:34.889542 138281463293504 feature_extraction.cc:275] Camera: #33 - SIMPLE_RADIAL\n",
560
+ "I20260112 15:58:34.889551 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
561
+ "I20260112 15:58:34.889564 138281463293504 feature_extraction.cc:282] Features: 3114 (SIFT)\n",
562
+ "I20260112 15:58:35.603434 138281463293504 feature_extraction.cc:260] Processed file [36/52]\n",
563
+ "I20260112 15:58:35.603486 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453740954.png\n",
564
+ "I20260112 15:58:35.603493 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
565
+ "I20260112 15:58:35.603499 138281463293504 feature_extraction.cc:275] Camera: #35 - SIMPLE_RADIAL\n",
566
+ "I20260112 15:58:35.603506 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
567
+ "I20260112 15:58:35.603520 138281463293504 feature_extraction.cc:282] Features: 2146 (SIFT)\n",
568
+ "I20260112 15:58:35.841722 138281463293504 feature_extraction.cc:260] Processed file [37/52]\n",
569
+ "I20260112 15:58:35.841774 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453745156.png\n",
570
+ "I20260112 15:58:35.841782 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
571
+ "I20260112 15:58:35.841788 138281463293504 feature_extraction.cc:275] Camera: #36 - SIMPLE_RADIAL\n",
572
+ "I20260112 15:58:35.841796 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
573
+ "I20260112 15:58:35.841809 138281463293504 feature_extraction.cc:282] Features: 3020 (SIFT)\n",
574
+ "I20260112 15:58:38.459461 138281463293504 feature_extraction.cc:260] Processed file [38/52]\n",
575
+ "I20260112 15:58:38.460112 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453753160.png\n",
576
+ "I20260112 15:58:38.460234 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
577
+ "I20260112 15:58:38.460311 138281463293504 feature_extraction.cc:275] Camera: #37 - SIMPLE_RADIAL\n",
578
+ "I20260112 15:58:38.460370 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
579
+ "I20260112 15:58:38.460529 138281463293504 feature_extraction.cc:282] Features: 3455 (SIFT)\n",
580
+ "I20260112 15:58:38.946302 138281463293504 feature_extraction.cc:260] Processed file [39/52]\n",
581
+ "I20260112 15:58:38.946920 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453765165.png\n",
582
+ "I20260112 15:58:38.948481 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
583
+ "I20260112 15:58:38.948509 138281463293504 feature_extraction.cc:275] Camera: #40 - SIMPLE_RADIAL\n",
584
+ "I20260112 15:58:38.948519 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
585
+ "I20260112 15:58:38.948537 138281463293504 feature_extraction.cc:282] Features: 914 (SIFT)\n",
586
+ "I20260112 15:58:39.206731 138281463293504 feature_extraction.cc:260] Processed file [40/52]\n",
587
+ "I20260112 15:58:39.207329 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453756762.png\n",
588
+ "I20260112 15:58:39.207417 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
589
+ "I20260112 15:58:39.207428 138281463293504 feature_extraction.cc:275] Camera: #38 - SIMPLE_RADIAL\n",
590
+ "I20260112 15:58:39.207436 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
591
+ "I20260112 15:58:39.207451 138281463293504 feature_extraction.cc:282] Features: 3341 (SIFT)\n",
592
+ "I20260112 15:58:39.457441 138281463293504 feature_extraction.cc:260] Processed file [41/52]\n",
593
+ "I20260112 15:58:39.457500 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453759963.png\n",
594
+ "I20260112 15:58:39.457508 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
595
+ "I20260112 15:58:39.457514 138281463293504 feature_extraction.cc:275] Camera: #39 - SIMPLE_RADIAL\n",
596
+ "I20260112 15:58:39.457521 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
597
+ "I20260112 15:58:39.457535 138281463293504 feature_extraction.cc:282] Features: 2544 (SIFT)\n",
598
+ "I20260112 15:58:41.905434 138281463293504 feature_extraction.cc:260] Processed file [42/52]\n",
599
+ "I20260112 15:58:41.906331 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453774370.png\n",
600
+ "I20260112 15:58:41.906416 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
601
+ "I20260112 15:58:41.906672 138281463293504 feature_extraction.cc:275] Camera: #41 - SIMPLE_RADIAL\n",
602
+ "I20260112 15:58:41.906742 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
603
+ "I20260112 15:58:41.906961 138281463293504 feature_extraction.cc:282] Features: 1301 (SIFT)\n",
604
+ "I20260112 15:58:42.735463 138281463293504 feature_extraction.cc:260] Processed file [43/52]\n",
605
+ "I20260112 15:58:42.735515 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453779372.png\n",
606
+ "I20260112 15:58:42.735523 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
607
+ "I20260112 15:58:42.735529 138281463293504 feature_extraction.cc:275] Camera: #42 - SIMPLE_RADIAL\n",
608
+ "I20260112 15:58:42.735537 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
609
+ "I20260112 15:58:42.735550 138281463293504 feature_extraction.cc:282] Features: 2205 (SIFT)\n",
610
+ "I20260112 15:58:43.132374 138281463293504 feature_extraction.cc:260] Processed file [44/52]\n",
611
+ "I20260112 15:58:43.133269 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453786375.png\n",
612
+ "I20260112 15:58:43.133394 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
613
+ "I20260112 15:58:43.133628 138281463293504 feature_extraction.cc:275] Camera: #44 - SIMPLE_RADIAL\n",
614
+ "I20260112 15:58:43.133857 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
615
+ "I20260112 15:58:43.133956 138281463293504 feature_extraction.cc:282] Features: 1376 (SIFT)\n",
616
+ "I20260112 15:58:43.140206 138281463293504 feature_extraction.cc:260] Processed file [45/52]\n",
617
+ "I20260112 15:58:43.140679 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453783374.png\n",
618
+ "I20260112 15:58:43.140960 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
619
+ "I20260112 15:58:43.143426 138281463293504 feature_extraction.cc:275] Camera: #43 - SIMPLE_RADIAL\n",
620
+ "I20260112 15:58:43.143471 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
621
+ "I20260112 15:58:43.143485 138281463293504 feature_extraction.cc:282] Features: 2433 (SIFT)\n",
622
+ "I20260112 15:58:45.897146 138281463293504 feature_extraction.cc:260] Processed file [46/52]\n",
623
+ "I20260112 15:58:45.898444 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453790978.png\n",
624
+ "I20260112 15:58:45.898521 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
625
+ "I20260112 15:58:45.898573 138281463293504 feature_extraction.cc:275] Camera: #45 - SIMPLE_RADIAL\n",
626
+ "I20260112 15:58:45.898669 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
627
+ "I20260112 15:58:45.898772 138281463293504 feature_extraction.cc:282] Features: 1774 (SIFT)\n",
628
+ "I20260112 15:58:46.464438 138281463293504 feature_extraction.cc:260] Processed file [47/52]\n",
629
+ "I20260112 15:58:46.464499 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453793579.png\n",
630
+ "I20260112 15:58:46.464507 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
631
+ "I20260112 15:58:46.464514 138281463293504 feature_extraction.cc:275] Camera: #46 - SIMPLE_RADIAL\n",
632
+ "I20260112 15:58:46.464521 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
633
+ "I20260112 15:58:46.464535 138281463293504 feature_extraction.cc:282] Features: 2368 (SIFT)\n",
634
+ "I20260112 15:58:46.685048 138281463293504 feature_extraction.cc:260] Processed file [48/52]\n",
635
+ "I20260112 15:58:46.685127 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453798181.png\n",
636
+ "I20260112 15:58:46.685134 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
637
+ "I20260112 15:58:46.685141 138281463293504 feature_extraction.cc:275] Camera: #47 - SIMPLE_RADIAL\n",
638
+ "I20260112 15:58:46.685201 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
639
+ "I20260112 15:58:46.685215 138281463293504 feature_extraction.cc:282] Features: 2249 (SIFT)\n",
640
+ "I20260112 15:58:46.858438 138281463293504 feature_extraction.cc:260] Processed file [49/52]\n",
641
+ "I20260112 15:58:46.858498 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453801783.png\n",
642
+ "I20260112 15:58:46.858505 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
643
+ "I20260112 15:58:46.858511 138281463293504 feature_extraction.cc:275] Camera: #48 - SIMPLE_RADIAL\n",
644
+ "I20260112 15:58:46.858519 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
645
+ "I20260112 15:58:46.858532 138281463293504 feature_extraction.cc:282] Features: 2189 (SIFT)\n",
646
+ "I20260112 15:58:49.002276 138281463293504 feature_extraction.cc:260] Processed file [50/52]\n",
647
+ "I20260112 15:58:49.002340 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453805788.png\n",
648
+ "I20260112 15:58:49.002347 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
649
+ "I20260112 15:58:49.002354 138281463293504 feature_extraction.cc:275] Camera: #49 - SIMPLE_RADIAL\n",
650
+ "I20260112 15:58:49.002361 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
651
+ "I20260112 15:58:49.002375 138281463293504 feature_extraction.cc:282] Features: 2190 (SIFT)\n",
652
+ "I20260112 15:58:49.332376 138281463293504 feature_extraction.cc:260] Processed file [51/52]\n",
653
+ "I20260112 15:58:49.332454 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453862225.png\n",
654
+ "I20260112 15:58:49.332463 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
655
+ "I20260112 15:58:49.332470 138281463293504 feature_extraction.cc:275] Camera: #50 - SIMPLE_RADIAL\n",
656
+ "I20260112 15:58:49.332478 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
657
+ "I20260112 15:58:49.332492 138281463293504 feature_extraction.cc:282] Features: 1250 (SIFT)\n",
658
+ "I20260112 15:58:49.641517 138281463293504 feature_extraction.cc:260] Processed file [52/52]\n",
659
+ "I20260112 15:58:49.641570 138281463293504 feature_extraction.cc:263] Name: stairs_split_2_1710453871430.png\n",
660
+ "I20260112 15:58:49.641578 138281463293504 feature_extraction.cc:272] Dimensions: 1280 x 1024\n",
661
+ "I20260112 15:58:49.641585 138281463293504 feature_extraction.cc:275] Camera: #51 - SIMPLE_RADIAL\n",
662
+ "I20260112 15:58:49.641593 138281463293504 feature_extraction.cc:278] Focal Length: 1536.00px\n",
663
+ "I20260112 15:58:49.641608 138281463293504 feature_extraction.cc:282] Features: 2526 (SIFT)\n",
664
+ "I20260112 15:58:49.659250 138281496864320 timer.cc:90] Elapsed time: 0.806 [minutes]\n",
665
+ "I20260112 15:58:49.669833 138281496864320 misc.cc:44] \n",
666
+ "==============================================================================\n",
667
+ "Feature matching & geometric verification\n",
668
+ "==============================================================================\n",
669
+ "I20260112 15:58:49.670163 138281463293504 sift.cc:1452] Creating SIFT CPU feature matcher\n",
670
+ "I20260112 15:58:49.670228 138281471686208 sift.cc:1452] Creating SIFT CPU feature matcher\n",
671
+ "I20260112 15:58:49.670281 138281480078912 sift.cc:1452] Creating SIFT CPU feature matcher\n",
672
+ "I20260112 15:58:49.670317 138281612211776 sift.cc:1452] Creating SIFT CPU feature matcher\n",
673
+ "I20260112 15:58:49.670771 138281496864320 pairing.cc:180] Generating exhaustive image pairs...\n",
674
+ "I20260112 15:58:49.670794 138281496864320 pairing.cc:213] Processing block [1/2, 1/2]\n"
675
+ ]
676
+ },
677
+ {
678
+ "name": "stdout",
679
+ "output_type": "stream",
680
+ "text": [
681
+ " Extracted features from 51 images\n",
682
+ "Running COLMAP feature matching...\n"
683
+ ]
684
+ },
685
+ {
686
+ "name": "stderr",
687
+ "output_type": "stream",
688
+ "text": [
689
+ "I20260112 15:59:03.071126 138281496864320 feature_matching.cc:117] in 13.400s\n",
690
+ "I20260112 15:59:03.071220 138281496864320 pairing.cc:213] Processing block [1/2, 2/2]\n",
691
+ "I20260112 15:59:03.071233 138281496864320 feature_matching.cc:117] in 0.000s\n",
692
+ "I20260112 15:59:03.071243 138281496864320 pairing.cc:213] Processing block [2/2, 1/2]\n"
693
+ ]
694
+ },
695
+ {
696
+ "name": "stdout",
697
+ "output_type": "stream",
698
+ "text": [
699
+ " Feature matching completed\n",
700
+ "Running COLMAP mapper...\n",
701
+ " Mapper failed: incremental_mapping(): incompatible function arguments. The following argument types are supported:\n",
702
+ " 1. (database_path: str, image_path: str, output_path: str, options: pycolmap._core.IncrementalPipelineOptions = IncrementalPipelineOptions(), input_path: str = '', initial_image_pair_callback: collections.abc.Callable[[], None] = None, next_image_callback: collections.abc.Callable[[], None] = None) -> dict[int, pycolmap._core.Reconstruction]\n",
703
+ "\n",
704
+ "Invoked with: kwargs: database_path='/kaggle/working/result/features/stairs/database.db', image_path='/kaggle/input/image-matching-challenge-2025/test/stairs', output_path='/kaggle/working/result/reconstructions/stairs', options=IncrementalMapperOptions(init_min_num_inliers=100, init_max_error=4.0, init_max_forward_motion=0.95, init_min_tri_angle=16.0, init_max_reg_trials=2, abs_pose_max_error=12.0, abs_pose_min_num_inliers=30, abs_pose_min_inlier_ratio=0.25, abs_pose_refine_focal_length=True, abs_pose_refine_extra_params=True, ba_local_num_images=6, ba_local_min_tri_angle=6.0, ba_global_ignore_redundant_points3D=False, ba_global_prune_points_min_coverage_gain=0.05, min_focal_length_ratio=0.1, max_focal_length_ratio=10.0, max_extra_param=1.0, filter_max_reproj_error=4.0, filter_min_tri_angle=1.5, max_reg_trials=3, fix_existing_frames=False, constant_rigs=set(), constant_cameras=set(), num_threads=-1, random_seed=-1, image_selection_method=ImageSelectionMethod.MIN_UNCERTAINTY)\n",
705
+ "✅ Completed stairs\n",
706
+ "\n",
707
+ "Creating submission: /kaggle/working/submission.csv\n",
708
+ " Saved 73 entries\n",
709
+ "\n",
710
+ "======================================================================\n",
711
+ "FINAL RESULTS\n",
712
+ "======================================================================\n",
713
+ "Total images: 73\n",
714
+ "Registered with pose: 0\n",
715
+ "Success rate: 0.0%\n",
716
+ "Output: /kaggle/working/submission.csv\n",
717
+ "\n",
718
+ "First 5 predictions:\n",
719
+ "image_id,dataset,scene,image,rotation_matrix,translation_vector\n",
720
+ "ETs_another_et_another_et001.png_public,ETs,outliers,another_et_another_et001.png,nan;nan;nan;nan;nan;nan;nan;nan;nan,nan;nan;nan\n",
721
+ "ETs_another_et_another_et002.png_public,ETs,outliers,another_et_another_et002.png,nan;nan;nan;nan;nan;nan;nan;nan;nan,nan;nan;nan\n",
722
+ "ETs_another_et_another_et003.png_public,ETs,outliers,another_et_another_et003.png,nan;nan;nan;nan;nan;nan;nan;nan;nan,nan;nan;nan\n",
723
+ "ETs_another_et_another_et004.png_public,ETs,outliers,another_et_another_et004.png,nan;nan;nan;nan;nan;nan;nan;nan;nan,nan;nan;nan\n",
724
+ "ETs_another_et_another_et005.png_public,ETs,outliers,another_et_another_et005.png,nan;nan;nan;nan;nan;nan;nan;nan;nan,nan;nan;nan\n",
725
+ "\n",
726
+ "======================================================================\n",
727
+ "✅ PIPELINE COMPLETED SUCCESSFULLY!\n",
728
+ "======================================================================\n"
729
+ ]
730
+ },
731
+ {
732
+ "name": "stderr",
733
+ "output_type": "stream",
734
+ "text": [
735
+ "I20260112 15:59:03.862444 138281496864320 feature_matching.cc:117] in 0.791s\n",
736
+ "I20260112 15:59:03.862492 138281496864320 pairing.cc:213] Processing block [2/2, 2/2]\n",
737
+ "I20260112 15:59:03.862503 138281496864320 feature_matching.cc:117] in 0.000s\n",
738
+ "I20260112 15:59:03.862511 138281496864320 timer.cc:90] Elapsed time: 0.237 [minutes]\n",
739
+ "Traceback (most recent call last):\n",
740
+ " File \"/tmp/ipykernel_13/3332504298.py\", line 204, in run_colmap_mapper\n",
741
+ " maps = pycolmap.incremental_mapping(\n",
742
+ " ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n",
743
+ "TypeError: incremental_mapping(): incompatible function arguments. The following argument types are supported:\n",
744
+ " 1. (database_path: str, image_path: str, output_path: str, options: pycolmap._core.IncrementalPipelineOptions = IncrementalPipelineOptions(), input_path: str = '', initial_image_pair_callback: collections.abc.Callable[[], None] = None, next_image_callback: collections.abc.Callable[[], None] = None) -> dict[int, pycolmap._core.Reconstruction]\n",
745
+ "\n",
746
+ "Invoked with: kwargs: database_path='/kaggle/working/result/features/stairs/database.db', image_path='/kaggle/input/image-matching-challenge-2025/test/stairs', output_path='/kaggle/working/result/reconstructions/stairs', options=IncrementalMapperOptions(init_min_num_inliers=100, init_max_error=4.0, init_max_forward_motion=0.95, init_min_tri_angle=16.0, init_max_reg_trials=2, abs_pose_max_error=12.0, abs_pose_min_num_inliers=30, abs_pose_min_inlier_ratio=0.25, abs_pose_refine_focal_length=True, abs_pose_refine_extra_params=True, ba_local_num_images=6, ba_local_min_tri_angle=6.0, ba_global_ignore_redundant_points3D=False, ba_global_prune_points_min_coverage_gain=0.05, min_focal_length_ratio=0.1, max_focal_length_ratio=10.0, max_extra_param=1.0, filter_max_reproj_error=4.0, filter_min_tri_angle=1.5, max_reg_trials=3, fix_existing_frames=False, constant_rigs=set(), constant_cameras=set(), num_threads=-1, random_seed=-1, image_selection_method=ImageSelectionMethod.MIN_UNCERTAINTY)\n"
747
+ ]
748
+ }
749
+ ],
750
+ "source": [
751
+ "\"\"\"\n",
752
+ "IMC2025 COLMAP Pipeline v4 - Working Version\n",
753
+ "pycolmap API完全対応版\n",
754
+ "\n",
755
+ "【使い方】\n",
756
+ "1. Kaggle Settings → Internet を ON\n",
757
+ "2. !pip install pycolmap\n",
758
+ "3. Kernel → Restart\n",
759
+ "4. このコードを実行\n",
760
+ "\"\"\"\n",
761
+ "\n",
762
+ "import os\n",
763
+ "import sys\n",
764
+ "import glob\n",
765
+ "import shutil\n",
766
+ "import numpy as np\n",
767
+ "import pandas as pd\n",
768
+ "from pathlib import Path\n",
769
+ "from tqdm import tqdm\n",
770
+ "import dataclasses\n",
771
+ "from collections import defaultdict\n",
772
+ "import subprocess\n",
773
+ "from PIL import Image\n",
774
+ "import traceback\n",
775
+ "import gc\n",
776
+ "\n",
777
+ "# pycolmapのインポート\n",
778
+ "try:\n",
779
+ " import pycolmap\n",
780
+ " PYCOLMAP_AVAILABLE = True\n",
781
+ " print(\"✓ pycolmap is available\")\n",
782
+ "except ImportError:\n",
783
+ " PYCOLMAP_AVAILABLE = False\n",
784
+ " print(\"✗ pycolmap not found\")\n",
785
+ " print(\"\\nPlease install: !pip install pycolmap\")\n",
786
+ " sys.exit(1)\n",
787
+ "\n",
788
+ "# =========================================================\n",
789
+ "# 設定\n",
790
+ "# =========================================================\n",
791
+ "\n",
792
+ "@dataclasses.dataclass\n",
793
+ "class Prediction:\n",
794
+ " image_id: str | None\n",
795
+ " dataset: str\n",
796
+ " filename: str\n",
797
+ " map_size: int = -1\n",
798
+ " cluster_index: int | None = None\n",
799
+ " rotation: np.ndarray | None = None\n",
800
+ " translation: np.ndarray | None = None\n",
801
+ "\n",
802
+ "@dataclasses.dataclass\n",
803
+ "class DatasetParams:\n",
804
+ " dataset: str\n",
805
+ " feature_dir: str\n",
806
+ " images_dir: str\n",
807
+ " output_dir: str\n",
808
+ " database_path: str\n",
809
+ " predictions: list\n",
810
+ " filename_to_index: dict\n",
811
+ " images: list\n",
812
+ " models: dict = dataclasses.field(default_factory=dict)\n",
813
+ "\n",
814
+ "class CONFIG:\n",
815
+ " sift_num_features = 8192\n",
816
+ " sift_edge_threshold = 10\n",
817
+ " sift_peak_threshold = 0.0066\n",
818
+ " min_num_matches = 15\n",
819
+ " min_model_size = 3\n",
820
+ " max_num_models = 5\n",
821
+ " init_num_trials = 200\n",
822
+ " camera_model = \"SIMPLE_RADIAL\"\n",
823
+ "\n",
824
+ "# =========================================================\n",
825
+ "# ヘルパー関数\n",
826
+ "# =========================================================\n",
827
+ "\n",
828
+ "def setup_directories(base_dir, dataset):\n",
829
+ " \"\"\"ディレクトリ構造を設定\"\"\"\n",
830
+ " feature_dir = os.path.join(base_dir, 'features', dataset)\n",
831
+ " output_dir = os.path.join(base_dir, 'reconstructions', dataset)\n",
832
+ " database_path = os.path.join(feature_dir, 'database.db')\n",
833
+ " \n",
834
+ " for dir_path in [feature_dir, output_dir]:\n",
835
+ " if os.path.exists(dir_path):\n",
836
+ " shutil.rmtree(dir_path)\n",
837
+ " os.makedirs(dir_path, exist_ok=True)\n",
838
+ " \n",
839
+ " return feature_dir, output_dir, database_path\n",
840
+ "\n",
841
+ "def check_reconstruction_validity(reconstruction):\n",
842
+ " \"\"\"復元結果の妥当性をチェック\"\"\"\n",
843
+ " if reconstruction is None:\n",
844
+ " return False\n",
845
+ " \n",
846
+ " try:\n",
847
+ " if len(reconstruction.images) < CONFIG.min_model_size:\n",
848
+ " return False\n",
849
+ " if len(reconstruction.points3D) < 10:\n",
850
+ " return False\n",
851
+ " return True\n",
852
+ " except:\n",
853
+ " return False\n",
854
+ "\n",
855
+ "# =========================================================\n",
856
+ "# COLMAPパイプライン\n",
857
+ "# =========================================================\n",
858
+ "\n",
859
+ "def run_colmap_feature_extraction(images_dir, database_path):\n",
860
+ " \"\"\"COLMAP特徴抽出\"\"\"\n",
861
+ " print(\"Running COLMAP feature extraction...\")\n",
862
+ " \n",
863
+ " try:\n",
864
+ " # FeatureExtractionOptionsを使用（正しいクラス）\n",
865
+ " extraction_options = pycolmap.FeatureExtractionOptions()\n",
866
+ " \n",
867
+ " # SIFTパラメータの設定を試みる\n",
868
+ " try:\n",
869
+ " extraction_options.sift.max_num_features = CONFIG.sift_num_features\n",
870
+ " extraction_options.sift.edge_threshold = CONFIG.sift_edge_threshold\n",
871
+ " extraction_options.sift.peak_threshold = CONFIG.sift_peak_threshold\n",
872
+ " except AttributeError:\n",
873
+ " # バージョンによっては直接設定\n",
874
+ " pass\n",
875
+ " \n",
876
+ " # 特徴抽出実行\n",
877
+ " pycolmap.extract_features(\n",
878
+ " database_path=database_path,\n",
879
+ " image_path=images_dir,\n",
880
+ " camera_mode=pycolmap.CameraMode.AUTO,\n",
881
+ " camera_model=CONFIG.camera_model,\n",
882
+ " extraction_options=extraction_options\n",
883
+ " )\n",
884
+ " \n",
885
+ " # 画像数確認 - Databaseを直接開かずにファイルシステムで確認\n",
886
+ " image_files = [f for f in os.listdir(images_dir) \n",
887
+ " if f.lower().endswith(('.jpg', '.png', '.jpeg'))]\n",
888
+ " num_images = len(image_files)\n",
889
+ " \n",
890
+ " print(f\" Extracted features from {num_images} images\")\n",
891
+ " return num_images > 0\n",
892
+ " \n",
893
+ " except Exception as e:\n",
894
+ " print(f\" Feature extraction failed: {e}\")\n",
895
+ " traceback.print_exc()\n",
896
+ " return False\n",
897
+ "\n",
898
+ "def run_colmap_feature_matching(database_path):\n",
899
+ " \"\"\"COLMAP特徴マッチング\"\"\"\n",
900
+ " print(\"Running COLMAP feature matching...\")\n",
901
+ " \n",
902
+ " try:\n",
903
+ " # FeatureMatchingOptionsを使用（正しいクラス）\n",
904
+ " matching_options = pycolmap.FeatureMatchingOptions()\n",
905
+ " \n",
906
+ " # 利用可能な属性のみ設定\n",
907
+ " try:\n",
908
+ " # pycolmapのバージョンによって異なる属性\n",
909
+ " if hasattr(matching_options, 'max_num_matches'):\n",
910
+ " matching_options.max_num_matches = 32768\n",
911
+ " if hasattr(matching_options, 'max_error'):\n",
912
+ " matching_options.max_error = 4.0\n",
913
+ " except:\n",
914
+ " pass\n",
915
+ " \n",
916
+ " # Exhaustive matching - 正しいパラメータ名を使用\n",
917
+ " pycolmap.match_exhaustive(\n",
918
+ " database_path=database_path,\n",
919
+ " matching_options=matching_options\n",
920
+ " )\n",
921
+ " \n",
922
+ " print(\" Feature matching completed\")\n",
923
+ " return True\n",
924
+ " \n",
925
+ " except Exception as e:\n",
926
+ " print(f\" Feature matching failed: {e}\")\n",
927
+ " traceback.print_exc()\n",
928
+ " return False\n",
929
+ "\n",
930
+ "def run_colmap_mapper(database_path, images_dir, output_dir):\n",
931
+ " \"\"\"COLMAPマッパー（SfM）\"\"\"\n",
932
+ " print(\"Running COLMAP mapper...\")\n",
933
+ " \n",
934
+ " try:\n",
935
+ " # マッパーオプション - 正しい属性名を使用\n",
936
+ " mapper_options = pycolmap.IncrementalMapperOptions()\n",
937
+ " \n",
938
+ " # 利用可能な属性のみ設定\n",
939
+ " try:\n",
940
+ " if hasattr(mapper_options, 'min_model_size'):\n",
941
+ " mapper_options.min_model_size = CONFIG.min_model_size\n",
942
+ " if hasattr(mapper_options, 'max_num_models'):\n",
943
+ " mapper_options.max_num_models = CONFIG.max_num_models\n",
944
+ " if hasattr(mapper_options, 'num_threads'):\n",
945
+ " mapper_options.num_threads = -1\n",
946
+ " if hasattr(mapper_options, 'init_num_trials'):\n",
947
+ " mapper_options.init_num_trials = CONFIG.init_num_trials\n",
948
+ " if hasattr(mapper_options, 'extract_colors'):\n",
949
+ " mapper_options.extract_colors = False\n",
950
+ " except:\n",
951
+ " pass\n",
952
+ " \n",
953
+ " # SfM実行 - デフォルトオプションを使用\n",
954
+ " maps = pycolmap.incremental_mapping(\n",
955
+ " database_path=database_path,\n",
956
+ " image_path=images_dir,\n",
957
+ " output_path=output_dir,\n",
958
+ " options=mapper_options\n",
959
+ " )\n",
960
+ " \n",
961
+ " # 有効なモデルを抽出\n",
962
+ " valid_models = {}\n",
963
+ " if maps:\n",
964
+ " for idx, reconstruction in maps.items():\n",
965
+ " if check_reconstruction_validity(reconstruction):\n",
966
+ " model_dir = os.path.join(output_dir, str(idx))\n",
967
+ " os.makedirs(model_dir, exist_ok=True)\n",
968
+ " reconstruction.write(model_dir)\n",
969
+ " valid_models[model_dir] = reconstruction\n",
970
+ " \n",
971
+ " print(f\" Model {idx}: {len(reconstruction.images)} images, \"\n",
972
+ " f\"{len(reconstruction.points3D)} points\")\n",
973
+ " \n",
974
+ " if not valid_models:\n",
975
+ " print(\" No valid models created\")\n",
976
+ " \n",
977
+ " return valid_models\n",
978
+ " \n",
979
+ " except Exception as e:\n",
980
+ " print(f\" Mapper failed: {e}\")\n",
981
+ " traceback.print_exc()\n",
982
+ " return {}\n",
983
+ "\n",
984
+ "def run_bundle_adjustment(model_dir):\n",
985
+ " \"\"\"バンドル調整\"\"\"\n",
986
+ " try:\n",
987
+ " reconstruction = pycolmap.Reconstruction(model_dir)\n",
988
+ " ba_options = pycolmap.BundleAdjustmentOptions()\n",
989
+ " pycolmap.bundle_adjustment(reconstruction, ba_options)\n",
990
+ " reconstruction.write(model_dir)\n",
991
+ " return True\n",
992
+ " except Exception as e:\n",
993
+ " print(f\" Bundle adjustment warning: {e}\")\n",
994
+ " return False\n",
995
+ "\n",
996
+ "# =========================================================\n",
997
+ "# 結果処理\n",
998
+ "# =========================================================\n",
999
+ "\n",
1000
+ "def extract_poses_from_reconstruction(reconstruction):\n",
1001
+ " \"\"\"復元から姿勢を抽出\"\"\"\n",
1002
+ " poses = {}\n",
1003
+ " \n",
1004
+ " if reconstruction is None:\n",
1005
+ " return poses\n",
1006
+ " \n",
1007
+ " for image_id, image in reconstruction.images.items():\n",
1008
+ " try:\n",
1009
+ " filename = image.name\n",
1010
+ " cam_from_world = image.cam_from_world\n",
1011
+ " \n",
1012
+ " poses[filename] = {\n",
1013
+ " 'rotation': cam_from_world.rotation.matrix(),\n",
1014
+ " 'translation': cam_from_world.translation,\n",
1015
+ " 'camera_id': image.camera_id,\n",
1016
+ " 'num_points': len(image.points2D)\n",
1017
+ " }\n",
1018
+ " except Exception as e:\n",
1019
+ " print(f\" Warning: Failed to extract pose for {filename}\")\n",
1020
+ " \n",
1021
+ " return poses\n",
1022
+ "\n",
1023
+ "def update_predictions_with_poses(predictions, filename_to_index, reconstructions):\n",
1024
+ " \"\"\"予測を姿勢情報で更新\"\"\"\n",
1025
+ " updated_count = 0\n",
1026
+ " \n",
1027
+ " # 最大のモデルを優先\n",
1028
+ " sorted_recons = sorted(\n",
1029
+ " reconstructions.items(),\n",
1030
+ " key=lambda x: len(x[1].images),\n",
1031
+ " reverse=True\n",
1032
+ " )\n",
1033
+ " \n",
1034
+ " for recon_path, reconstruction in sorted_recons:\n",
1035
+ " if reconstruction is None:\n",
1036
+ " continue\n",
1037
+ " \n",
1038
+ " poses = extract_poses_from_reconstruction(reconstruction)\n",
1039
+ " \n",
1040
+ " for filename, pose_info in poses.items():\n",
1041
+ " if filename in filename_to_index:\n",
1042
+ " idx = filename_to_index[filename]\n",
1043
+ " \n",
1044
+ " # より大きなモデルで更新\n",
1045
+ " if predictions[idx].map_size < len(reconstruction.images):\n",
1046
+ " predictions[idx].rotation = pose_info['rotation']\n",
1047
+ " predictions[idx].translation = pose_info['translation']\n",
1048
+ " predictions[idx].map_size = len(reconstruction.images)\n",
1049
+ " predictions[idx].cluster_index = 0\n",
1050
+ " updated_count += 1\n",
1051
+ " \n",
1052
+ " print(f\" Updated {updated_count} predictions with poses\")\n",
1053
+ " return predictions\n",
1054
+ "\n",
1055
+ "# =========================================================\n",
1056
+ "# メインパイプライン\n",
1057
+ "# =========================================================\n",
1058
+ "\n",
1059
+ "def process_single_dataset(dataset_params):\n",
1060
+ " \"\"\"単一データセットを処理\"\"\"\n",
1061
+ " print(f\"\\n{'='*60}\")\n",
1062
+ " print(f\"Processing: {dataset_params.dataset}\")\n",
1063
+ " print(f\"Images: {len(dataset_params.images)}\")\n",
1064
+ " print('='*60)\n",
1065
+ " \n",
1066
+ " # ステップ1: 特徴抽出\n",
1067
+ " if not run_colmap_feature_extraction(\n",
1068
+ " dataset_params.images_dir, \n",
1069
+ " dataset_params.database_path\n",
1070
+ " ):\n",
1071
+ " return dataset_params\n",
1072
+ " \n",
1073
+ " # ステップ2: 特徴マッチング\n",
1074
+ " if not run_colmap_feature_matching(dataset_params.database_path):\n",
1075
+ " return dataset_params\n",
1076
+ " \n",
1077
+ " # ステップ3: SfM復元\n",
1078
+ " models = run_colmap_mapper(\n",
1079
+ " dataset_params.database_path,\n",
1080
+ " dataset_params.images_dir,\n",
1081
+ " dataset_params.output_dir\n",
1082
+ " )\n",
1083
+ " \n",
1084
+ " if not models:\n",
1085
+ " return dataset_params\n",
1086
+ " \n",
1087
+ " # ステップ4: バンドル調整\n",
1088
+ " for model_dir in models.keys():\n",
1089
+ " run_bundle_adjustment(model_dir)\n",
1090
+ " \n",
1091
+ " # ステップ5: 予測を更新\n",
1092
+ " dataset_params.models = models\n",
1093
+ " dataset_params.predictions = update_predictions_with_poses(\n",
1094
+ " dataset_params.predictions,\n",
1095
+ " dataset_params.filename_to_index,\n",
1096
+ " models\n",
1097
+ " )\n",
1098
+ " \n",
1099
+ " # サマリー\n",
1100
+ " registered = sum(1 for p in dataset_params.predictions if p.rotation is not None)\n",
1101
+ " print(f\"\\nResult: {registered}/{len(dataset_params.predictions)} images registered \"\n",
1102
+ " f\"({registered/len(dataset_params.predictions)*100:.1f}%)\")\n",
1103
+ " \n",
1104
+ " return dataset_params\n",
1105
+ "\n",
1106
+ "# =========================================================\n",
1107
+ "# サブミッション作成\n",
1108
+ "# =========================================================\n",
1109
+ "\n",
1110
+ "def create_submission(predictions, output_path, is_train=False):\n",
1111
+ " \"\"\"サブミッションファイルを作成\"\"\"\n",
1112
+ " print(f\"\\nCreating submission: {output_path}\")\n",
1113
+ " \n",
1114
+ " def array_to_str(array):\n",
1115
+ " return ';'.join([f\"{x:.09f}\" for x in array.flatten()])\n",
1116
+ " \n",
1117
+ " def none_to_str(n):\n",
1118
+ " return ';'.join(['nan'] * n)\n",
1119
+ " \n",
1120
+ " with open(output_path, 'w') as f:\n",
1121
+ " if is_train:\n",
1122
+ " f.write('dataset,scene,image,rotation_matrix,translation_vector\\n')\n",
1123
+ " for pred in predictions:\n",
1124
+ " scene = 'outliers' if pred.rotation is None else f'scene_{pred.cluster_index or 0}'\n",
1125
+ " rotation = none_to_str(9) if pred.rotation is None else array_to_str(pred.rotation)\n",
1126
+ " translation = none_to_str(3) if pred.translation is None else array_to_str(pred.translation)\n",
1127
+ " f.write(f'{pred.dataset},{scene},{pred.filename},{rotation},{translation}\\n')\n",
1128
+ " else:\n",
1129
+ " f.write('image_id,dataset,scene,image,rotation_matrix,translation_vector\\n')\n",
1130
+ " for pred in predictions:\n",
1131
+ " scene = 'outliers' if pred.rotation is None else f'scene_{pred.cluster_index or 0}'\n",
1132
+ " rotation = none_to_str(9) if pred.rotation is None else array_to_str(pred.rotation)\n",
1133
+ " translation = none_to_str(3) if pred.translation is None else array_to_str(pred.translation)\n",
1134
+ " f.write(f'{pred.image_id},{pred.dataset},{scene},{pred.filename},{rotation},{translation}\\n')\n",
1135
+ " \n",
1136
+ " print(f\" Saved {len(predictions)} entries\")\n",
1137
+ "\n",
1138
+ "# =========================================================\n",
1139
+ "# メイン実行\n",
1140
+ "# =========================================================\n",
1141
+ "\n",
1142
+ "def main():\n",
1143
+ " \"\"\"メイン実行関数\"\"\"\n",
1144
+ " \n",
1145
+ " # 環境設定\n",
1146
+ " is_train = False\n",
1147
+ " data_dir = '/kaggle/input/image-matching-challenge-2025'\n",
1148
+ " workdir = '/kaggle/working/result'\n",
1149
+ " \n",
1150
+ " os.makedirs(workdir, exist_ok=True)\n",
1151
+ " \n",
1152
+ " # CSVパス\n",
1153
+ " csv_path = os.path.join(\n",
1154
+ " data_dir, \n",
1155
+ " 'train_labels.csv' if is_train else 'sample_submission.csv'\n",
1156
+ " )\n",
1157
+ " \n",
1158
+ " # データ読み込み\n",
1159
+ " print(\"Loading competition data...\")\n",
1160
+ " competition_data = pd.read_csv(csv_path)\n",
1161
+ " \n",
1162
+ " # テスト用フィルタリング\n",
1163
+ " if not is_train and len(competition_data) == 1945:\n",
1164
+ " test_datasets = [\"ETs\", \"stairs\"]\n",
1165
+ " competition_data = competition_data[\n",
1166
+ " competition_data[\"dataset\"].isin(test_datasets)\n",
1167
+ " ]\n",
1168
+ " print(f\"Testing with {len(competition_data)} entries from {test_datasets}\")\n",
1169
+ " \n",
1170
+ " # 予測オブジェクト作成\n",
1171
+ " dataset_predictions = defaultdict(list)\n",
1172
+ " for _, row in competition_data.iterrows():\n",
1173
+ " pred = Prediction(\n",
1174
+ " image_id=None if is_train else row.image_id,\n",
1175
+ " dataset=row.dataset,\n",
1176
+ " filename=row.image\n",
1177
+ " )\n",
1178
+ " dataset_predictions[row.dataset].append(pred)\n",
1179
+ " \n",
1180
+ " print(f\"Found {len(dataset_predictions)} datasets\")\n",
1181
+ " \n",
1182
+ " # 各データセットを処理\n",
1183
+ " all_predictions = []\n",
1184
+ " \n",
1185
+ " for dataset, predictions in dataset_predictions.items():\n",
1186
+ " print(f\"\\n{'='*70}\")\n",
1187
+ " print(f\"DATASET: {dataset}\")\n",
1188
+ " print('='*70)\n",
1189
+ " \n",
1190
+ " try:\n",
1191
+ " # 画像ディレクトリ\n",
1192
+ " images_dir = os.path.join(\n",
1193
+ " data_dir, \n",
1194
+ " 'train' if is_train else 'test', \n",
1195
+ " dataset\n",
1196
+ " )\n",
1197
+ " \n",
1198
+ " if not os.path.exists(images_dir):\n",
1199
+ " print(f\"Directory not found: {images_dir}\")\n",
1200
+ " all_predictions.extend(predictions)\n",
1201
+ " continue\n",
1202
+ " \n",
1203
+ " # 有効な画像をフィルタ\n",
1204
+ " valid_predictions = []\n",
1205
+ " valid_images = []\n",
1206
+ " \n",
1207
+ " for pred in predictions:\n",
1208
+ " img_path = os.path.join(images_dir, pred.filename)\n",
1209
+ " if os.path.exists(img_path):\n",
1210
+ " valid_predictions.append(pred)\n",
1211
+ " valid_images.append(img_path)\n",
1212
+ " \n",
1213
+ " if len(valid_images) < CONFIG.min_model_size:\n",
1214
+ " print(f\"Skipping: only {len(valid_images)} images\")\n",
1215
+ " all_predictions.extend(predictions)\n",
1216
+ " continue\n",
1217
+ " \n",
1218
+ " # ファイル��インデックス\n",
1219
+ " filename_to_index = {\n",
1220
+ " p.filename: idx \n",
1221
+ " for idx, p in enumerate(valid_predictions)\n",
1222
+ " }\n",
1223
+ " \n",
1224
+ " # ディレクトリ設定\n",
1225
+ " feature_dir, output_dir, database_path = setup_directories(\n",
1226
+ " workdir, dataset\n",
1227
+ " )\n",
1228
+ " \n",
1229
+ " # パラメータ作成\n",
1230
+ " params = DatasetParams(\n",
1231
+ " dataset=dataset,\n",
1232
+ " feature_dir=feature_dir,\n",
1233
+ " images_dir=images_dir,\n",
1234
+ " output_dir=output_dir,\n",
1235
+ " database_path=database_path,\n",
1236
+ " predictions=valid_predictions,\n",
1237
+ " filename_to_index=filename_to_index,\n",
1238
+ " images=valid_images\n",
1239
+ " )\n",
1240
+ " \n",
1241
+ " # 処理実行\n",
1242
+ " result = process_single_dataset(params)\n",
1243
+ " all_predictions.extend(result.predictions)\n",
1244
+ " \n",
1245
+ " print(f\"✅ Completed {dataset}\")\n",
1246
+ " \n",
1247
+ " # メモリクリーンアップ\n",
1248
+ " gc.collect()\n",
1249
+ " \n",
1250
+ " except Exception as e:\n",
1251
+ " print(f\"❌ Failed: {e}\")\n",
1252
+ " traceback.print_exc()\n",
1253
+ " all_predictions.extend(predictions)\n",
1254
+ " \n",
1255
+ " # サブミッション作成\n",
1256
+ " submission_file = '/kaggle/working/submission.csv'\n",
1257
+ " create_submission(all_predictions, submission_file, is_train)\n",
1258
+ " \n",
1259
+ " # 最終サマリー\n",
1260
+ " print(f\"\\n{'='*70}\")\n",
1261
+ " print(\"FINAL RESULTS\")\n",
1262
+ " print('='*70)\n",
1263
+ " \n",
1264
+ " registered = sum(1 for p in all_predictions if p.rotation is not None)\n",
1265
+ " total = len(all_predictions)\n",
1266
+ " \n",
1267
+ " print(f\"Total images: {total}\")\n",
1268
+ " print(f\"Registered with pose: {registered}\")\n",
1269
+ " print(f\"Success rate: {registered/total*100:.1f}%\")\n",
1270
+ " print(f\"Output: {submission_file}\")\n",
1271
+ " \n",
1272
+ " # サンプル表示\n",
1273
+ " print(\"\\nFirst 5 predictions:\")\n",
1274
+ " with open(submission_file, 'r') as f:\n",
1275
+ " for i, line in enumerate(f):\n",
1276
+ " if i < 6:\n",
1277
+ " print(line.strip())\n",
1278
+ " \n",
1279
+ " return all_predictions\n",
1280
+ "\n",
1281
+ "# =========================================================\n",
1282
+ "# 実行\n",
1283
+ "# =========================================================\n",
1284
+ "\n",
1285
+ "if __name__ == \"__main__\":\n",
1286
+ " print(\"=\" * 70)\n",
1287
+ " print(\"IMC2025 COLMAP PIPELINE v4\")\n",
1288
+ " print(\"=\" * 70)\n",
1289
+ " \n",
1290
+ " try:\n",
1291
+ " results = main()\n",
1292
+ " \n",
1293
+ " print(\"\\n\" + \"=\" * 70)\n",
1294
+ " print(\"✅ PIPELINE COMPLETED SUCCESSFULLY!\")\n",
1295
+ " print(\"=\" * 70)\n",
1296
+ " \n",
1297
+ " # 成功した画像の詳細\n",
1298
+ " successful = [p for p in results if p.rotation is not None]\n",
1299
+ " if successful:\n",
1300
+ " print(f\"\\nSuccessfully registered {len(successful)} images\")\n",
1301
+ " for i, pred in enumerate(successful[:5]):\n",
1302
+ " print(f\" {pred.filename}: model_size={pred.map_size}\")\n",
1303
+ " \n",
1304
+ " except KeyboardInterrupt:\n",
1305
+ " print(\"\\nInterrupted by user\")\n",
1306
+ " sys.exit(1)\n",
1307
+ " except Exception as e:\n",
1308
+ " print(f\"\\nFatal error: {e}\")\n",
1309
+ " traceback.print_exc()\n",
1310
+ " sys.exit(1)"
1311
+ ]
1312
+ }
1313
+ ],
1314
+ "metadata": {
1315
+ "kaggle": {
1316
+ "accelerator": "none",
1317
+ "dataSources": [
1318
+ {
1319
+ "databundleVersionId": 11655853,
1320
+ "sourceId": 91498,
1321
+ "sourceType": "competition"
1322
+ }
1323
+ ],
1324
+ "isGpuEnabled": false,
1325
+ "isInternetEnabled": true,
1326
+ "language": "python",
1327
+ "sourceType": "notebook"
1328
+ },
1329
+ "kernelspec": {
1330
+ "display_name": "Python 3",
1331
+ "language": "python",
1332
+ "name": "python3"
1333
+ },
1334
+ "language_info": {
1335
+ "codemirror_mode": {
1336
+ "name": "ipython",
1337
+ "version": 3
1338
+ },
1339
+ "file_extension": ".py",
1340
+ "mimetype": "text/x-python",
1341
+ "name": "python",
1342
+ "nbconvert_exporter": "python",
1343
+ "pygments_lexer": "ipython3",
1344
+ "version": "3.11.13"
1345
+ },
1346
+ "papermill": {
1347
+ "default_parameters": {},
1348
+ "duration": 85.367019,
1349
+ "end_time": "2026-01-12T15:59:04.468755",
1350
+ "environment_variables": {},
1351
+ "exception": null,
1352
+ "input_path": "__notebook__.ipynb",
1353
+ "output_path": "__notebook__.ipynb",
1354
+ "parameters": {},
1355
+ "start_time": "2026-01-12T15:57:39.101736",
1356
+ "version": "2.6.0"
1357
+ }
1358
+ },
1359
+ "nbformat": 4,
1360
+ "nbformat_minor": 5
1361
+ }