Hugging Face's logo

Datasets:
imageomics
/
kabr-worked-examples

Tasks:
Video Classification
Modalities:
Tabular
Text
Video
Languages:
English
Size:
10K - 100K
ArXiv:
Tags:
zebra
giraffe
plains zebra
Grevy's zebra
video
animal behavior
DOI:
License:
Dataset card Data Studio Files
xet
Community
6
jennamk14 commited on
Commit
4a89330
·
1 Parent(s): 4e93be2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +741 -3
README.md CHANGED
@@ -1,3 +1,741 @@
1
- ---
2
- license: cc0-1.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc0-1.0
3
+ task_categories:
4
+ - video-classification
5
+ tags:
6
+ - zebra
7
+ - giraffe
8
+ - plains zebra
9
+ - Grevy's zebra
10
+ - video
11
+ - animal behavior
12
+ - behavior recognition
13
+ - annotation
14
+ - annotated video
15
+ - conservation
16
+ - drone
17
+ - UAV
18
+ - imbalanced
19
+ - Kenya
20
+ - Mpala Research Centre
21
+ language:
22
+ - en
23
+ pretty_name: kabr-worked-examples
24
+ size_categories:
25
+ - 1K<n<10K
26
+ ---
27
+
28
+ # Dataset Card for kabr-worked-example
29
+
30
+ ## Table of Contents
31
+ - [Dataset Description](#dataset-description)
32
+ - [Dataset Details](#dataset-details)
33
+ - [Dataset Description](#dataset-description-1)
34
+ - [Session Summary](#session-summary)
35
+ - [Dataset Structure](#dataset-structure)
36
+ - [What each file/folder is for](#what-each-filefolder-is-for)
37
+ - [Data instances](#data-instances)
38
+ - [Data fields](#data-fields)
39
+ - [A. Detections (CVAT "tracks" XML)](#a-detections-cvat-tracks-xml)
40
+ - [B. Behavior CSV (auto labels; one file per source video)](#b-behavior-csv-auto-labels-one-file-per-source-video)
41
+ - [C. Mini-scene metadata JSON (per source video)](#c-mini-scene-metadata-json-per-source-video)
42
+ - [D. Actions folder (per-clip predictions; if present)](#d-actions-folder-per-clip-predictions-if-present)
43
+ - [E. Telemetry CSV (Airdata export)](#e-telemetry-csv-airdata-export)
44
+ - [Dataset Creation](#dataset-creation)
45
+ - [Curation Rationale](#curation-rationale)
46
+ - [Source Data](#source-data)
47
+ - [Data Collection and Processing](#data-collection-and-processing)
48
+ - [Who are the source data producers?](#who-are-the-source-data-producers)
49
+ - [Annotations](#annotations)
50
+ - [Annotation process](#annotation-process)
51
+ - [Who are the annotators?](#who-are-the-annotators)
52
+ - [Personal and Sensitive Information](#personal-and-sensitive-information)
53
+ - [Considerations for Using the Data](#considerations-for-using-the-data)
54
+ - [References](#references)
55
+ - [Citation](#citation)
56
+ - [Contributions](#contributions)
57
+
58
+
59
+ ## Dataset Details
60
+ Manually annotated bounding box detections, mini-scenes, behavior annotations, and associated telemetry
61
+ for three sessions used for kabr-tools case studies.
62
+
63
+ ### Dataset Description
64
+
65
+ - **Curated by:** Alison Zhong and Jenna Kline
66
+ - **Homepage:**
67
+ - **Repository:** https://github.com/Imageomics/kabr-tools
68
+ - **Paper:** kabr-tools (manuscript in preparation)
69
+
70
+ Annotations were created to evaluate the kabr-tools pipeline and conduct case studies on Grevy's landscape of fear and inter-species spatial distribution. Annotations include manual detections and tracks, mini-scenes cut from source videos, behavior annotations from an X3D action recognition model, and associated drone telemetry data. The detections contains bounding box coordinates, image file names, and class labels for each annotated animal. Annotations were created using CVAT to manually draw bounding boxes around animals in a selection of raw drone videos. The annotations were then exported as xml files and used to create the provided mini-scenes. The KABR X3D model was used to label the mini-scenes with predicted behaviors. Telemetry data was exported from Airdata.
71
+
72
+ ### Session Summary
73
+
74
+ | Session | Date Collected | Demographic Information and Habitat | Video File IDs in Session |
75
+ |---------|---------------|--------------|---------|
76
+ | `1` | 2023-01-18 | 2 Adult male Grevy's zebra in an open plain| DJI_0068, DJI_0069, DJI_0070, DJI_0071 |
77
+ | `2` | 2023-01-20 | 5 Grevy's zebra in a semi-open habitat along a roadway| DJI_0142, DJI_0143, DJI_0144, DJI_0145, DJI_0146, DJI_0147 |
78
+ | `3` | 2023-01-21 | Mixed herd of 3 reticulated giraffe, 2 plains zebras and 11 Grevy's zebras in a closed habitat with dense vegetation near Mo Kenya| DJI_0206, DJI_0208, DJI_0210, DJI_0211 |
79
+
80
+ ## Dataset Structure
81
+ ```text
82
+
83
+ ├── behavior
84
+ │ ├── 18_01_2023_session_7-DJI_0068.csv
85
+ │ ├── 18_01_2023_session_7-DJI_0069.csv
86
+ │ ├── 18_01_2023_session_7-DJI_0070.csv
87
+ │ ├── 18_01_2023_session_7-DJI_0071.csv
88
+ │ ├── 20_01_2023_session_3-DJI_0142.csv
89
+ │ ├── 20_01_2023_session_3-DJI_0143.csv
90
+ │ ├── 20_01_2023_session_3-DJI_0144.csv
91
+ │ ├── 20_01_2023_session_3-DJI_0145.csv
92
+ │ ├── 20_01_2023_session_3-DJI_0146.csv
93
+ │ ├── 20_01_2023_session_3-DJI_0147.csv
94
+ │ ├── 21_01_2023_session_5-DJI_0206.csv
95
+ │ ├── 21_01_2023_session_5-DJI_0208.csv
96
+ │ ├── 21_01_2023_session_5-DJI_0210.csv
97
+ │ ├── 21_01_2023_session_5-DJI_0211.csv
98
+ │ └── 21_01_2023_session_5-DJI_0212.csv
99
+ ├── detections
100
+ │ ├── 18_01_2023_session_7-DJI_0068.xml
101
+ │ ├── 18_01_2023_session_7-DJI_0069.xml
102
+ │ ├── 18_01_2023_session_7-DJI_0070.xml
103
+ │ ├── 18_01_2023_session_7-DJI_0071.xml
104
+ │ ├── 20_01_2023_session_3-DJI_0142.xml
105
+ │ ├── 20_01_2023_session_3-DJI_0143.xml
106
+ │ ├── 20_01_2023_session_3-DJI_0144.xml
107
+ │ ├── 20_01_2023_session_3-DJI_0145.xml
108
+ │ ├── 20_01_2023_session_3-DJI_0146.xml
109
+ │ ├── 20_01_2023_session_3-DJI_0147.xml
110
+ │ ├── 21_01_2023_session_5-DJI_0206.xml
111
+ │ ├── 21_01_2023_session_5-DJI_0208.xml
112
+ │ ├── 21_01_2023_session_5-DJI_0210.xml
113
+ │ ├── 21_01_2023_session_5-DJI_0211.xml
114
+ │ └── 21_01_2023_session_5-DJI_0212.xml
115
+ ├── mini_scenes
116
+ │ ├── 18_01_2023_session_7-DJI_0068
117
+ │ │ ├── 0.mp4
118
+ │ │ ├── 1.mp4
119
+ │ │ ├── DJI_0068.mp4
120
+ │ │ └── metadata
121
+ │ │ ├── DJI_0068.jpg
122
+ │ │ ├── DJI_0068_metadata.json
123
+ │ │ └── DJI_0068_tracks.xml
124
+ │ ├── 18_01_2023_session_7-DJI_0069
125
+ │ │ ├── 0.mp4
126
+ │ │ ├── 1.mp4
127
+ │ │ ├── DJI_0069.mp4
128
+ │ │ └── metadata
129
+ │ │ ├── DJI_0069.jpg
130
+ │ │ ├── DJI_0069_metadata.json
131
+ │ │ └── DJI_0069_tracks.xml
132
+ │ ├── 18_01_2023_session_7-DJI_0070
133
+ │ │ ├── 0.mp4
134
+ │ │ ├── 1.mp4
135
+ │ │ ├── DJI_0070.mp4
136
+ │ │ └── metadata
137
+ │ │ ├── DJI_0070.jpg
138
+ │ │ ├── DJI_0070_metadata.json
139
+ │ │ └── DJI_0070_tracks.xml
140
+ │ ├── 18_01_2023_session_7-DJI_0071
141
+ │ │ ├── 0.mp4
142
+ │ │ ├── 1.mp4
143
+ │ │ ├── DJI_0071.mp4
144
+ │ │ └── metadata
145
+ │ │ ├── DJI_0071.jpg
146
+ │ │ ├── DJI_0071_metadata.json
147
+ │ │ └── DJI_0071_tracks.xml
148
+ │ ├── 20_01_2023_session_3-DJI_0142
149
+ │ │ ├── 0.mp4
150
+ │ │ ├── 10.mp4
151
+ │ │ ├── 11.mp4
152
+ │ │ ├── 1.mp4
153
+ │ │ ├── 2.mp4
154
+ │ │ ├── 3.mp4
155
+ │ │ ├── 4.mp4
156
+ │ │ ├── 5.mp4
157
+ │ │ ├── 6.mp4
158
+ │ │ ├── 7.mp4
159
+ │ │ ├── 8.mp4
160
+ │ │ ├── 9.mp4
161
+ │ │ ├── actions
162
+ │ │ ├── DJI_0142.mp4
163
+ │ │ └── metadata
164
+ │ │ ├── DJI_0142.jpg
165
+ │ │ ├── DJI_0142_metadata.json
166
+ │ │ └── DJI_0142_tracks.xml
167
+ │ ├── 20_01_2023_session_3-DJI_0143
168
+ │ │ ├── 0.mp4
169
+ │ │ ├── 1.mp4
170
+ │ │ ├── 2.mp4
171
+ │ │ ├── 3.mp4
172
+ │ │ ├── 4.mp4
173
+ │ │ ├── actions
174
+ │ │ ├── DJI_0143.mp4
175
+ │ │ └── metadata
176
+ │ │ ├── DJI_0143.jpg
177
+ │ │ ├── DJI_0143_metadata.json
178
+ │ │ └── DJI_0143_tracks.xml
179
+ │ ├── 20_01_2023_session_3-DJI_0144
180
+ │ │ ├── 0.mp4
181
+ │ │ ├── 1.mp4
182
+ │ │ ├── 2.mp4
183
+ │ │ ├── 3.mp4
184
+ │ │ ├── 4.mp4
185
+ │ │ ├── 5.mp4
186
+ │ │ ├── actions
187
+ │ │ ├── DJI_0144.mp4
188
+ │ │ └── metadata
189
+ │ │ ├── DJI_0144.jpg
190
+ │ │ ├── DJI_0144_metadata.json
191
+ │ │ └── DJI_0144_tracks.xml
192
+ │ ├── 20_01_2023_session_3-DJI_0145
193
+ │ │ ├── 0.mp4
194
+ │ │ ├── 1.mp4
195
+ │ │ ├── 2.mp4
196
+ │ │ ├── 3.mp4
197
+ │ │ ├── 4.mp4
198
+ │ │ ├── actions
199
+ │ │ ├── DJI_0145.mp4
200
+ │ │ └── metadata
201
+ │ │ ├── DJI_0145.jpg
202
+ │ │ ├── DJI_0145_metadata.json
203
+ │ │ └── DJI_0145_tracks.xml
204
+ │ ├── 20_01_2023_session_3-DJI_0146
205
+ │ │ ├── 0.mp4
206
+ │ │ ├── 1.mp4
207
+ │ │ ├── 2.mp4
208
+ │ │ ├── 3.mp4
209
+ │ │ ├── 4.mp4
210
+ │ │ ├── 5.mp4
211
+ │ │ ├── actions
212
+ │ │ ├── DJI_0146.mp4
213
+ │ │ └── metadata
214
+ │ │ ├── DJI_0146.jpg
215
+ │ │ ├── DJI_0146_metadata.json
216
+ │ │ └── DJI_0146_tracks.xml
217
+ │ ├── 21_01_2023_session_5-DJI_0206
218
+ │ │ ├── 0.mp4
219
+ │ │ ├── 10.mp4
220
+ │ │ ├── 11.mp4
221
+ │ │ ├── 12.mp4
222
+ │ │ ├── 13.mp4
223
+ │ │ ├── 14.mp4
224
+ │ │ ├── 15.mp4
225
+ │ │ ├── 16.mp4
226
+ │ │ ├── 17.mp4
227
+ │ │ ├── 18.mp4
228
+ │ │ ├── 19.mp4
229
+ │ │ ├── 1.mp4
230
+ │ │ ├── 20.mp4
231
+ │ │ ├── 21.mp4
232
+ │ │ ├── 22.mp4
233
+ │ │ ├── 23.mp4
234
+ │ │ ├── 24.mp4
235
+ │ │ ├── 25.mp4
236
+ │ │ ├── 26.mp4
237
+ │ │ ├── 27.mp4
238
+ │ │ ├── 28.mp4
239
+ │ │ ├── 29.mp4
240
+ │ │ ├── 2.mp4
241
+ │ │ ├── 30.mp4
242
+ │ │ ├── 31.mp4
243
+ │ │ ├── 3.mp4
244
+ │ │ ├── 4.mp4
245
+ │ │ ├── 5.mp4
246
+ │ │ ├── 6.mp4
247
+ │ │ ├── 7.mp4
248
+ │ │ ├── 8.mp4
249
+ │ │ ├── 9.mp4
250
+ │ │ ├── DJI_0206.mp4
251
+ │ │ └── metadata
252
+ │ │ ├── DJI_0206.jpg
253
+ │ │ ├── DJI_0206_metadata.json
254
+ │ │ └── DJI_0206_tracks.xml
255
+ │ ├── 21_01_2023_session_5-DJI_0208
256
+ │ │ ├── 0.mp4
257
+ │ │ ├── 10.mp4
258
+ │ │ ├── 11.mp4
259
+ │ │ ├── 12.mp4
260
+ │ │ ├── 13.mp4
261
+ │ │ ├── 14.mp4
262
+ │ │ ├── 15.mp4
263
+ │ │ ├── 16.mp4
264
+ │ │ ├── 17.mp4
265
+ │ │ ├── 18.mp4
266
+ │ │ ├── 19.mp4
267
+ │ │ ├── 1.mp4
268
+ │ │ ├── 20.mp4
269
+ │ │ ├── 21.mp4
270
+ │ │ ├── 22.mp4
271
+ │ │ ├── 23.mp4
272
+ │ │ ├── 24.mp4
273
+ │ │ ├── 25.mp4
274
+ │ │ ├── 26.mp4
275
+ │ │ ├── 27.mp4
276
+ │ │ ├── 28.mp4
277
+ │ │ ├── 29.mp4
278
+ │ │ ├── 2.mp4
279
+ │ │ ├── 30.mp4
280
+ │ │ ├── 31.mp4
281
+ │ │ ├── 32.mp4
282
+ │ │ ├── 33.mp4
283
+ │ │ ├── 34.mp4
284
+ │ │ ├── 35.mp4
285
+ │ │ ├── 36.mp4
286
+ │ │ ├── 37.mp4
287
+ │ │ ├── 38.mp4
288
+ │ │ ├── 39.mp4
289
+ │ │ ├── 3.mp4
290
+ │ │ ├── 40.mp4
291
+ │ │ ├── 41.mp4
292
+ │ │ ├── 42.mp4
293
+ │ │ ├── 43.mp4
294
+ │ │ ├── 44.mp4
295
+ │ │ ├── 45.mp4
296
+ │ │ ├── 46.mp4
297
+ │ │ ├── 47.mp4
298
+ │ │ ├── 48.mp4
299
+ │ │ ├── 49.mp4
300
+ │ │ ├── 4.mp4
301
+ │ │ ├── 50.mp4
302
+ │ │ ├── 51.mp4
303
+ │ │ ├── 52.mp4
304
+ │ │ ├── 53.mp4
305
+ │ │ ├── 54.mp4
306
+ │ │ ├── 55.mp4
307
+ │ │ ├── 5.mp4
308
+ │ │ ├── 6.mp4
309
+ │ │ ├── 7.mp4
310
+ │ │ ├── 8.mp4
311
+ │ │ ├── 9.mp4
312
+ │ │ ├── DJI_0208.mp4
313
+ │ │ └── metadata
314
+ │ │ ├── DJI_0208.jpg
315
+ │ │ ├── DJI_0208_metadata.json
316
+ │ │ └── DJI_0208_tracks.xml
317
+ │ ├── 21_01_2023_session_5-DJI_0210
318
+ │ │ ├── 0.mp4
319
+ │ │ ├── 10.mp4
320
+ │ │ ├── 11.mp4
321
+ │ │ ├── 12.mp4
322
+ │ │ ├── 13.mp4
323
+ │ │ ├── 14.mp4
324
+ │ │ ├── 15.mp4
325
+ │ │ ├── 16.mp4
326
+ │ │ ├── 17.mp4
327
+ │ │ ├── 18.mp4
328
+ │ │ ├── 19.mp4
329
+ │ │ ├── 1.mp4
330
+ │ │ ├── 20.mp4
331
+ │ │ ├── 21.mp4
332
+ │ │ ├── 22.mp4
333
+ │ │ ├── 23.mp4
334
+ │ │ ├── 24.mp4
335
+ │ │ ├── 2.mp4
336
+ │ │ ├── 3.mp4
337
+ │ │ ├── 4.mp4
338
+ │ │ ├── 5.mp4
339
+ │ │ ├── 6.mp4
340
+ │ │ ├── 7.mp4
341
+ │ │ ├── 8.mp4
342
+ │ │ ├── 9.mp4
343
+ │ │ ├── DJI_0210.mp4
344
+ │ │ └── metadata
345
+ │ │ ├── DJI_0210.jpg
346
+ │ │ ├── DJI_0210_metadata.json
347
+ │ │ └── DJI_0210_tracks.xml
348
+ │ ├── 21_01_2023_session_5-DJI_0211
349
+ │ │ ├── 0.mp4
350
+ │ │ ├── 10.mp4
351
+ │ │ ├── 11.mp4
352
+ │ │ ├── 12.mp4
353
+ │ │ ├── 13.mp4
354
+ │ │ ├── 14.mp4
355
+ │ │ ├── 15.mp4
356
+ │ │ ├── 16.mp4
357
+ │ │ ├── 17.mp4
358
+ │ │ ├── 18.mp4
359
+ │ │ ├── 19.mp4
360
+ │ │ ├── 1.mp4
361
+ │ │ ├── 20.mp4
362
+ │ │ ├── 21.mp4
363
+ │ │ ├── 22.mp4
364
+ │ │ ├── 23.mp4
365
+ │ │ ├── 24.mp4
366
+ │ │ ├── 25.mp4
367
+ │ │ ├── 26.mp4
368
+ │ │ ├── 27.mp4
369
+ │ │ ├── 28.mp4
370
+ │ │ ├── 29.mp4
371
+ │ │ ├── 2.mp4
372
+ │ │ ├── 30.mp4
373
+ │ │ ├── 31.mp4
374
+ │ │ ├── 32.mp4
375
+ │ │ ├── 33.mp4
376
+ │ │ ├── 3.mp4
377
+ │ │ ├── 4.mp4
378
+ │ │ ├── 5.mp4
379
+ │ │ ├── 6.mp4
380
+ │ │ ├── 7.mp4
381
+ │ │ ├── 8.mp4
382
+ │ │ ├── 9.mp4
383
+ │ │ ├── DJI_0211.mp4
384
+ │ │ └── metadata
385
+ │ │ ├── DJI_0211.jpg
386
+ │ │ ├── DJI_0211_metadata.json
387
+ │ │ └── DJI_0211_tracks.xml
388
+ │ └── 21_01_2023_session_5-DJI_0212
389
+ │ ├── 0.mp4
390
+ │ ├── 10.mp4
391
+ │ ├── 11.mp4
392
+ │ ├── 12.mp4
393
+ │ ├── 13.mp4
394
+ │ ├── 14.mp4
395
+ │ ├── 1.mp4
396
+ │ ├── 2.mp4
397
+ │ ├── 3.mp4
398
+ │ ├── 4.mp4
399
+ │ ├── 5.mp4
400
+ │ ├── 6.mp4
401
+ │ ├── 7.mp4
402
+ │ ├── 8.mp4
403
+ │ ├── 9.mp4
404
+ │ ├── DJI_0212.mp4
405
+ │ └── metadata
406
+ │ ├── DJI_0212.jpg
407
+ │ ├── DJI_0212_metadata.json
408
+ │ └── DJI_0212_tracks.xml
409
+ ├── README.md
410
+ └── telemetry
411
+ ├── Jan-18th-2023-12-47PM-Flight-Airdata.csv
412
+ ├── Jan-20th-2023-12-58PM-Flight-Airdata.csv
413
+ └── Jan-21st-2023-02-49PM-Flight-Airdata.csv
414
+
415
+ ```
416
+
417
+ ## What each file/folder is for
418
+
419
+ | Path / Pattern | Purpose |
420
+ |---|---|
421
+ | `detections/*.xml` | **Manual detections/tracks** per source video (CVAT “tracks” XML). One `<track>` per animal across frames; used to cut mini-scenes. |
422
+ | `mini_scenes/<video_id>/DJI_XXXX.mp4` | The **source video** referenced by detections for that `<video_id>`. |
423
+ | `mini_scenes/<video_id>/<k>.mp4` | **Mini-scenes** (short clips) cut from the source video based on detection tracks (`0.mp4`, `1.mp4`, …). |
424
+ | `mini_scenes/<video_id>/metadata/DJI_XXXX_tracks.xml` | Copy of the **CVAT tracks** used to generate the mini-scenes (provenance). |
425
+ | `mini_scenes/<video_id>/metadata/DJI_XXXX_metadata.json` | **Video-level metadata** (session/date, FPS, resolution, timing, etc.). |
426
+ | `mini_scenes/<video_id>/metadata/DJI_XXXX.jpg` | **Thumbnail/keyframe** for quick preview. |
427
+ | `mini_scenes/<video_id>/actions/` | **Per-clip auto behavior labels** from the X3D action model (CSV or JSON; presence varies by video). |
428
+ | `behavior/*.csv` | **Per-video roll-ups** of X3D behavior predictions. One row per mini-scene clip with label + references (and optional timing/confidence). |
429
+ | `telemetry/*Flight-Airdata.csv` | **Drone flight logs** (Airdata export) for the corresponding sessions (timing, altitude, battery, etc.). |
430
+ | `README.md` | Repository-level notes and usage tips. |
431
+
432
+
433
+ ## Data instances
434
+
435
+ - **Detection instance (XML):** one `<track>` spans frames; each `<box>` is a frame-level bounding box with coordinates and flags.
436
+ - **Mini-scene instance (MP4):** a short clip indexed by file name (`k.mp4`) under `mini_scenes/<video_id>/`.
437
+ - **Behavior instance (CSV row):** one mini-scene with **X3D-predicted behavior** and references to the clip (plus optional confidence/timing).
438
+ - **Telemetry instance (CSV row):** one flight-log record from Airdata with timestamped vehicle context.
439
+
440
+
441
+ ## Data fields
442
+
443
+ ### A. Detections (CVAT “tracks” XML)
444
+
445
+ | Element / Attribute | Type | Example | Meaning |
446
+ | --- | --- | --- | --- |
447
+ | `/annotations/version` | string | `1.1` | Annotation file/schema version. |
448
+ | `/annotations/track@id` | integer | `0` | Unique id for a tracked object. |
449
+ | `/annotations/track@label` | string | `Grevy` | Class/species label. |
450
+ | `/annotations/track@source` | string | `manual` | How the annotation was created. |
451
+ | `/annotations/track/box@frame` | int (0-based) | `0,1,2,…` | Frame index. |
452
+ | `/annotations/track/box@outside` | enum {`0`,`1`} | `0` | `0` present; `1` not visible. |
453
+ | `/annotations/track/box@occluded`| enum {`0`,`1`} | `0` | Occlusion flag. |
454
+ | `/annotations/track/box@keyframe`| enum {`0`,`1`} | `1` | Keyframe marker. |
455
+ | `/annotations/track/box@xtl` | float (px) | `2342.00` | X of top-left. |
456
+ | `/annotations/track/box@ytl` | float (px) | `2427.00` | Y of top-left. |
457
+ | `/annotations/track/box@xbr` | float (px) | `2530.00` | X of bottom-right. |
458
+ | `/annotations/track/box@ybr` | float (px) | `2623.00` | Y of bottom-right. |
459
+ | `/annotations/track/box@z_order` | integer | `0` | Drawing order. |
460
+
461
+
462
+ ### B. Behavior CSV (auto labels; one file per source video)
463
+
464
+ > **Note:** Column names may vary slightly by export; use the header in each CSV as ground truth.
465
+
466
+ | Column (typical) | Example | Meaning |
467
+ |---|---|---|
468
+ | `clip_path` or `clip_id` | `mini_scenes/21_01_2023_session_5-DJI_0208/33.mp4` | Relative path to the mini-scene clip. |
469
+ | `source_video` | `DJI_0208.mp4` | Name of the parent/source video. |
470
+ | `video_id` | `21_01_2023_session_5-DJI_0208` | Folder/video identifier. |
471
+ | `clip_index` | `33` | Index of the clip within the video folder. |
472
+ | `behavior` | `walking` | X3D-predicted action/behavior label. |
473
+ | `confidence` | `0.92` | Model confidence/probability (if provided). |
474
+ | `start_frame` | `1234` | First frame of the segment (if provided). |
475
+ | `end_frame` | `1450` | Last frame of the segment (if provided). |
476
+ | `start_time` | `00:00:41.2` | Segment start time (if provided). |
477
+ | `end_time` | `00:00:48.8` | Segment end time (if provided). |
478
+ | `species` | `Grevy` | Species label (if propagated/available). |
479
+ | `notes` | `—` | Free-text notes or flags (optional). |
480
+ | `model` | `x3d` | Model identifier used to label. |
481
+ | `model_version` | `x3d_m` | Specific checkpoint/version tag (optional). |
482
+
483
+
484
+ ### C. Mini-scene metadata JSON (per source video)
485
+
486
+ > **Typical keys** (presence may vary):
487
+
488
+ | Key | Example | Meaning |
489
+ |---|---|---|
490
+ | `video_id` | `21_01_2023_session_5-DJI_0208` | Folder/video identifier. |
491
+ | `source_video` | `DJI_0208.mp4` | Original MP4 filename. |
492
+ | `session_date` | `2023-01-21` | Capture date. |
493
+ | `session_id` | `session_5` | Field session tag. |
494
+ | `fps` | `29.97` | Frames per second. |
495
+ | `resolution` | `[3840, 2160]` | Width × height (px). |
496
+ | `duration_s` | `123.45` | Video duration (seconds). |
497
+ | `timezone` | `Africa/Nairobi` | Local timezone of recording. |
498
+ | `generator` | `mini_scene_cutter@<git-sha>` | Tool/commit that wrote the metadata. |
499
+ | `tracks_xml` | `DJI_0208_tracks.xml` | Provenance link to the CVAT tracks file. |
500
+
501
+
502
+ ### D. Actions folder (per-clip predictions; if present)
503
+
504
+ - **CSV format (typical columns):** `clip_index`, `clip_path`, `behavior`, `confidence`, `model`, `model_version`.
505
+ - **JSON format (typical fields):** object per clip with `index`, `path`, `label`, `score`, `model`, `model_version`.
506
+
507
+
508
+ ### E. Telemetry CSV (Airdata export)
509
+
510
+ > Columns depend on Airdata export settings; common fields include:
511
+
512
+ | Column (common) | Example | Meaning |
513
+ |---|---|---|
514
+ | `UTC Timestamp` | `2023-01-21 12:49:07` | Log timestamp (UTC). |
515
+ | `Latitude` / `Longitude` | `0.28123`, `37.12345` | Aircraft location. |
516
+ | `Altitude (m)` | `68.2` | Altitude above takeoff or MSL (per export). |
517
+ | `AGL (m)` | `47.9` | Above-ground level (if provided). |
518
+ | `Speed (m/s)` | `9.4` | Horizontal speed. |
519
+ | `Heading (deg)` | `135` | Yaw/heading. |
520
+ | `Battery (%)` | `54` | Remaining battery percentage. |
521
+ | `Flight Mode` | `P-GPS` | Autopilot mode. |
522
+ | `Distance (m)` | `122.5` | Distance from home point. |
523
+
524
+
525
+
526
+ ## Dataset Creation
527
+
528
+ ### Curation Rationale
529
+ Created to evaluate kabr-tools pipeline and conduct case studies on Grevy's landscape of fear and inter-species spatial distribution.
530
+
531
+ ### Source Data
532
+
533
+ <!-- This section describes the source data (e.g., news text and headlines, social media posts, translated sentences, ...). As well as an original source it was created from (e.g., sampling from Zenodo records, compiling images from different aggregators, etc.) -->
534
+
535
+ #### Data Collection and Processing
536
+ Data collected at Mpala Research Centre, Kenya, in January 2023. The data was collected using a DJI Air 2S drone and manually annotated using CVAT. The annotations were exported as xml files.
537
+
538
+ #### Who are the source data producers?
539
+ See citation for kabr-full-video dataset - https://huggingface.co/datasets/imageomics/KABR-full-videos
540
+
541
+
542
+ ### Annotations
543
+ <!--
544
+ If the dataset contains annotations which are not part of the initial data collection, use this section to describe them.
545
+
546
+ Ex: We standardized the taxonomic labels provided by the various data sources to conform to a uniform 7-rank Linnean structure. (Then, under annotation process, describe how this was done: Our sources used different names for the same kingdom (both _Animalia_ and _Metazoa_), so we chose one for all (_Animalia_). -->
547
+
548
+ #### Annotation process
549
+ CVAT was used to manually annotate the bounding boxes around animals in the videos. The annotations were then exported as xml files to create mini-scenes using tracks_extractor.py. The mini-scenes were then labeled with predicted behaviors using the KABR X3D action recognition model using the miniscene2behavior.py.
550
+
551
+ #### Who are the annotators?
552
+ Alison Zhong and Jenna Kline
553
+
554
+ ### Personal and Sensitive Information
555
+ People trimmed from the videos before annotation.
556
+ Endangered species are included in the dataset, but no personal or sensitive information is included.
557
+
558
+
559
+ ## Considerations for Using the Data
560
+
561
+ ### Intended Use Cases
562
+
563
+ This dataset serves as a **worked example** for the kabr-tools pipeline and is specifically designed for:
564
+ - **Pipeline demonstration**: Showing complete end-to-end processing from raw videos to behavioral annotations
565
+ - **Method validation**: Evaluating automated detection and behavior recognition against manual annotations
566
+ - **Case study research**: Supporting specific research questions on Grevy's zebra landscape of fear and inter-species spatial distribution
567
+ - **Educational purposes**: Teaching researchers how to use the kabr-tools pipeline with real data
568
+ - **Reproducibility**: Providing a reference implementation with known inputs and outputs
569
+
570
+ ### Important Data Considerations
571
+
572
+ **Limited scope**: This is a **demonstration dataset** with only 3 sessions and 15 video files, designed to illustrate methodology rather than provide comprehensive coverage.
573
+
574
+ **Session heterogeneity**: Each session represents distinctly different scenarios:
575
+ - **Session 1**: Minimal complexity (2 male Grevy's zebras, open habitat)
576
+ - **Session 2**: Moderate complexity (5 Grevy's zebras, semi-open roadway habitat)
577
+ - **Session 3**: High complexity (mixed species, dense vegetation, 16 total animals)
578
+
579
+ **Processing completeness**: Not all videos have complete processing outputs - some lack `actions/` folders, reflecting real-world pipeline execution variability.
580
+
581
+ **Annotation methodology**: Manual detections serve as ground truth, while behavior labels are X3D model predictions, not expert-validated behaviors.
582
+
583
+ ### Bias, Risks, and Limitations
584
+
585
+ **Sample size limitations**:
586
+ - Only 15 video files across 3 sessions
587
+ - Insufficient for statistical generalization
588
+ - Designed for demonstration, not comprehensive analysis
589
+
590
+ **Species representation bias**:
591
+ - Heavily weighted toward Grevy's zebras (endangered species focus)
592
+ - Giraffes only present in one session (Session 3)
593
+ - Plains zebras only in mixed-species context
594
+ - May not represent typical behavioral patterns for each species
595
+
596
+ **Habitat and temporal constraints**:
597
+ - Single location (Mpala Research Centre, Kenya)
598
+ - 3-day collection window (January 18-21, 2023)
599
+ - Limited environmental and seasonal variability
600
+ - Habitat types may not represent species' full range
601
+
602
+ **Technical processing limitations**:
603
+ - X3D behavior predictions are automated, not expert-validated
604
+ - Mini-scene extraction dependent on manual annotation quality
605
+ - Telemetry synchronization with video timestamps may require adjustment
606
+ - Some videos lack complete behavioral annotation outputs
607
+
608
+ **Methodological constraints**:
609
+ - Manual annotations by only 2 annotators (potential inter-annotator variability)
610
+ - CVAT tracking may have limitations in dense vegetation (Session 3)
611
+ - Behavior model trained on different dataset, may not generalize perfectly
612
+
613
+ ### Recommendations
614
+
615
+ **For pipeline evaluation and development**:
616
+ - Use manual detections in `detections/*.xml` as ground truth for automated detection validation
617
+ - Compare processing outputs across sessions to understand pipeline performance in different scenarios
618
+ - Use Session 1 (simple) for initial testing, Session 3 (complex) for stress testing
619
+ - Validate timestamp alignment between telemetry and video data before spatial analysis
620
+
621
+ **For case study research**:
622
+ - **Landscape of fear studies**: Focus on Grevy's zebra data from Sessions 1 and 2; use telemetry data to correlate spatial position with behaviors
623
+ - **Inter-species analysis**: Use Session 3 mixed-species data; consider habitat complexity when interpreting interactions
624
+ - Account for small sample sizes in statistical analyses and interpretation
625
+
626
+ **For educational use**:
627
+ - Start with Session 1 data for learning pipeline basics
628
+ - Progress through sessions in order of increasing complexity
629
+ - Use metadata files to understand processing provenance
630
+ - Examine both successful and incomplete processing examples
631
+
632
+ **Technical recommendations**:
633
+ - Verify file completeness before analysis (not all videos have `actions/` folders)
634
+ - Check CSV headers as column names may vary between exports
635
+ - Use metadata JSON files to understand video-specific processing parameters
636
+ - Cross-reference telemetry timestamps with video timing for spatial-behavioral analysis
637
+
638
+ **Data interpretation cautions**:
639
+ - Treat X3D behavior predictions as model outputs, not ground truth
640
+ - Consider habitat context when interpreting behavioral patterns
641
+ - Account for species-specific behavioral repertoires in analysis
642
+ - Use this dataset to understand methodology, not to draw broad ecological conclusions
643
+
644
+ ## References
645
+
646
+ - Original KABR dataset: [https://huggingface.co/datasets/imageomics/KABR](https://huggingface.co/datasets/imageomics/KABR)
647
+ - KABR raw videos: [https://huggingface.co/datasets/imageomics/KABR-full-videos](https://huggingface.co/datasets/imageomics/KABR-full-videos)
648
+ - kabr-tools repository: [https://github.com/Imageomics/kabr-tools](https://github.com/Imageomics/kabr-tools)
649
+ - Mpala Research Centre: [https://mpala.org/](https://mpala.org/)
650
+
651
+ ## Licensing Information
652
+
653
+ This dataset is dedicated to the public domain for the benefit of scientific pursuits under the CC0 1.0 Universal Public Domain Dedication. We ask that you cite the dataset and related publications using the citations below if you make use of it in your research.
654
+
655
+ ## Citation
656
+
657
+ **BibTeX:**
658
+
659
+ **Dataset**
660
+ ```
661
+ @misc{KABR_worked_example,
662
+ author = {Zhong, Alison and Kline, Jenna and Kholiavchenko, Maksim and Stevens, Sam and Sheets, Alec and Babu, Reshma and Banerji, Namrata and Campolongo, Elizabeth and Thompson, Matthew and Van Tiel, Nina and Miliko, Jackson and Rosser, Neil and Stewart, Charles and Berger-Wolf, Tanya and Rubenstein, Daniel},
663
+ title = {KABR Worked Example: Manually Annotated Detections and Behavioral Analysis for Kenyan Wildlife Pipeline Demonstration},
664
+ year = {2023},
665
+ url = {https://huggingface.co/datasets/imageomics/kabr-worked-example},
666
+ publisher = {Hugging Face}
667
+ }
668
+ ```
669
+
670
+ **Related Publications**
671
+ ```
672
+ @inproceedings{kholiavchenko2024kabr,
673
+ title={KABR: In-Situ Dataset for Kenyan Animal Behavior Recognition from Drone Videos},
674
+ author={Kholiavchenko, Maksim and Kline, Jenna and Ramirez, Michelle and Stevens, Sam and Sheets, Alec and Babu, Reshma and Banerji, Namrata and Campolongo, Elizabeth and Thompson, Matthew and Van Tiel, Nina and Miliko, Jackson and Bessa, Eduardo and Duporge, Isla and Berger-Wolf, Tanya and Rubenstein, Daniel and Stewart, Charles},
675
+ booktitle={Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision},
676
+ pages={31-40},
677
+ year={2024}
678
+ }
679
+ ```
680
+
681
+ **kabr-tools manuscript (in preparation)**
682
+ ```
683
+ @article{kabr_tools_manuscript,
684
+ title={kabr-tools: An Open-Source Pipeline for Automated Wildlife Behavior Analysis from Drone Videos},
685
+ author={Zhong, Alison and Kline, Jenna and [additional authors]},
686
+ journal={[Journal name]},
687
+ year={[Year]},
688
+ note={Manuscript in preparation}
689
+ }
690
+ ```
691
+
692
+ Please also cite the original data sources:
693
+ - Original KABR dataset: https://huggingface.co/datasets/imageomics/KABR
694
+ - KABR raw videos: https://huggingface.co/datasets/imageomics/KABR-full-videos
695
+
696
+ ## Contributions
697
+
698
+ This work was supported by the [Imageomics Institute](https://imageomics.org), which is funded by the US National Science Foundation's Harnessing the Data Revolution (HDR) program under [Award #2118240](https://www.nsf.gov/awardsearch/showAward?AWD_ID=2118240) (Imageomics: A New Frontier of Biological Information Powered by Knowledge-Guided Machine Learning). Additional support was provided by the [AI Institute for Intelligent Cyberinfrastructure with Computational Learning in the Environment (ICICLE)](https://icicle.osu.edu/), funded by the US National Science Foundation under [Award #2112606](https://www.nsf.gov/awardsearch/showAward?AWD_ID=2112606).
699
+
700
+ Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
701
+
702
+ The data was collected at the [Mpala Research Centre](https://mpala.org/) in Kenya, in accordance with Research License No. NACOSTI/P/22/18214. The data collection protocol adhered strictly to the guidelines set forth by the Institutional Animal Care and Use Committee under permission No. IACUC 1835F.
703
+
704
+ ### Dataset Creation Contributors
705
+
706
+ - **Data Collection**: Field team at Mpala Research Centre, Kenya
707
+ - **Manual Annotations**: Alison Zhong and Jenna Kline
708
+ - **Pipeline Development**: kabr-tools development team
709
+ - **Behavioral Analysis**: X3D model predictions using KABR-trained models
710
+ - **Data Curation**: Alison Zhong and Jenna Kline
711
+ - **Quality Assurance**: Imageomics Institute research team
712
+
713
+ ## Glossary
714
+
715
+ **Mini-scene**: Short video clips (typically 5-10 seconds) extracted from source videos, centered on individual animals based on tracking annotations.
716
+
717
+ **CVAT**: Computer Vision Annotation Tool - open-source software used for manual video annotation and object tracking.
718
+
719
+ **X3D**: 3D CNN architecture used for video-based action recognition, adapted for animal behavior classification in the KABR project.
720
+
721
+ **Track**: A sequence of bounding boxes following a single animal across multiple video frames.
722
+
723
+ **Telemetry**: Flight data recorded by the drone during video capture, including GPS coordinates, altitude, speed, and battery status.
724
+
725
+ **Session**: A discrete data collection period, typically representing one flight or filming session on a specific date.
726
+
727
+ ## More Information
728
+
729
+ For detailed usage instructions and code examples, see the [kabr-tools repository](https://github.com/Imageomics/kabr-tools).
730
+
731
+ For questions about the broader KABR project and related datasets, visit the [Imageomics Institute website](https://imageomics.org).
732
+
733
+ This dataset is part of a larger effort to develop automated methods for wildlife monitoring and conservation using computer vision and machine learning techniques.
734
+
735
+ ## Dataset Card Authors
736
+
737
+ Jenna Kline
738
+
739
+ ## Dataset Card Contact
740
+
741
+ kline dot 377 at osu dot edu