Datasets:

Modalities:
Image
Text
Languages:
English
ArXiv:
DOI:
License:
jennamk14 commited on
Commit
e154dfc
·
verified ·
1 Parent(s): 1d1908a

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +210 -46
README.md CHANGED
@@ -30,6 +30,8 @@ This is a dataset containing annotated video frames of giraffes, Grevy's zebras,
30
  - **Repository:** [https://github.com/Imageomics/wildwing](https://github.com/Imageomics/wildwing)
31
  - **Papers:**
32
 
 
 
33
  - [WildWing: An open-source, autonomous and affordable UAS for animal behaviour video monitoring](https://doi.org/10.1111/2041-210X.70018)
34
 
35
  - [Deep Dive KABR](https://doi.org/10.1007/s11042-024-20512-4)
@@ -38,7 +40,6 @@ This is a dataset containing annotated video frames of giraffes, Grevy's zebras,
38
 
39
 
40
 
41
-
42
  <!-- Provide a longer summary of what this dataset is. -->
43
  This dataset contains video frames collected as part of the [Kenyan Animal Behavior Recognition (KABR)](https://kabrdata.xyz/) project at the [Mpala Research Center](https://mpala.org/) in Kenya in January 2023.
44
 
@@ -47,14 +48,14 @@ Sessions 1 and 2 are part of the [original KABR data release](https://huggingfac
47
  The dataset consists of 104,062 frames. Each frame is accompanied by annotations in COCO format, indicating the presence of zebras and giraffes and their bounding boxes within the images. The annotations were completed manually by the dataset curator using [CVAT](https://www.cvat.ai/) and [kabr-tools](https://github.com/Imageomics/kabr-tools).
48
 
49
 
50
- | Session | Date Collected | Total Frames | Species |
51
- |---------|---------------|--------------|---------------------------------|
52
- | `session_1` | 2023-01-12 | 16,891 | Giraffe |
53
- | `session_2` | 2023-01-17 | 11,165 | Plains zebra |
54
- | `session_3` | 2023-01-18 | 17,940 | Grevy's zebra |
55
- | `session_4` | 2023-01-20 | 33,960 | Grevy's zebra |
56
- | `session_5` | 2023-01-21 | 24,106 | Giraffe, Plains and Grevy's zebras |
57
- | **Total Frames:** | | **104,062** | |
58
 
59
  This table shows the data collected at Mpala Research Center in Laikipia, Kenya, with session information, dates, frame counts, and primary species observed.
60
 
@@ -65,39 +66,212 @@ The dataset is intended for use in training and evaluating computer vision model
65
  ## Dataset Structure
66
  ```​
67
  /dataset/
68
- <session_1>/
69
- <classes.txt>
70
- <video_id_frame 0>.jpg
71
- <video_id_frame 0>.txt
72
- <video_id_frame 1>.jpg
73
- ...
74
- <video_id_frame n>.jpg
75
- <video_id_frame n>.txt
76
- <session_2>/
77
- <classes.txt>
78
- ...
79
- <session_3>/
80
- <classes.txt>
81
- ...
82
- <session_4>/
83
- <classes.txt>
84
- ...
85
- <session_5>
86
- <classes.txt>
87
- ...
88
- <video_id_frame n>.jpg
89
- <video_id_frame n>.txt
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
90
  ```
91
 
92
  ### Data Instances
93
  All images are named <video_id_frame>.jpg, each within a folder named for the date of the session. The annotations are in COCO format and are stored in a corresponding .txt file with the same name as the image.
94
 
 
 
95
 
96
  ### Data Fields
97
 
98
  **classes.txt**:
99
- - `0`: giraffe
100
- - `1`: zebra
 
 
101
 
102
  **frame_id.txt**:
103
  - `class`: Class of the object in the image (0 for zebra)
@@ -243,8 +417,10 @@ This dataset (the compilation) has been marked as dedicated to the public domain
243
  year={2024}
244
  }
245
  ```
 
 
246
  ```​
247
- article{kline2025wildwing,
248
  title={WildWing: An open-source, autonomous and affordable UAS for animal behaviour video monitoring},
249
  author={Kline, Jenna and Zhong, Alison and Irizarry, Kevyn and Stewart, Charles V and Stewart, Christopher and Rubenstein, Daniel I and Berger-Wolf, Tanya},
250
  journal={Methods in Ecology and Evolution},
@@ -254,18 +430,6 @@ article{kline2025wildwing,
254
  }
255
  ```
256
 
257
- ```​
258
- @misc{kline2025mmlamultienvironmentmultispecieslowaltitude,
259
- title={MMLA: Multi-Environment, Multi-Species, Low-Altitude Aerial Footage Dataset},
260
- author={Jenna Kline and Samuel Stevens and Guy Maalouf and Camille Rondeau Saint-Jean and Dat Nguyen Ngoc and Majid Mirmehdi and David Guerin and Tilo Burghardt and Elzbieta Pastucha and Blair Costelloe and Matthew Watson and Thomas Richardson and Ulrik Pagh Schultz Lundquist},
261
- year={2025},
262
- eprint={2504.07744},
263
- archivePrefix={arXiv},
264
- primaryClass={cs.CV},
265
- url={https://arxiv.org/abs/2504.07744},
266
- }
267
- ```
268
-
269
 
270
 
271
  ## Acknowledgements
 
30
  - **Repository:** [https://github.com/Imageomics/wildwing](https://github.com/Imageomics/wildwing)
31
  - **Papers:**
32
 
33
+ - [MMLA: Multi-Environment, Multi-Species, Low-Altitude Aerial Footage Dataset](https://arxiv.org/abs/2504.07744)
34
+
35
  - [WildWing: An open-source, autonomous and affordable UAS for animal behaviour video monitoring](https://doi.org/10.1111/2041-210X.70018)
36
 
37
  - [Deep Dive KABR](https://doi.org/10.1007/s11042-024-20512-4)
 
40
 
41
 
42
 
 
43
  <!-- Provide a longer summary of what this dataset is. -->
44
  This dataset contains video frames collected as part of the [Kenyan Animal Behavior Recognition (KABR)](https://kabrdata.xyz/) project at the [Mpala Research Center](https://mpala.org/) in Kenya in January 2023.
45
 
 
48
  The dataset consists of 104,062 frames. Each frame is accompanied by annotations in COCO format, indicating the presence of zebras and giraffes and their bounding boxes within the images. The annotations were completed manually by the dataset curator using [CVAT](https://www.cvat.ai/) and [kabr-tools](https://github.com/Imageomics/kabr-tools).
49
 
50
 
51
+ | Session | Date Collected | Total Frames | Species | Video File IDs |
52
+ |---------|---------------|--------------|---------|----------------|
53
+ | `session_1` | 2023-01-12 | 16,891 | Giraffe | DJI_0001, DJI_0002 |
54
+ | `session_2` | 2023-01-17 | 11,165 | Plains zebra | DJI_0005, DJI_0006 |
55
+ | `session_3` | 2023-01-18 | 17,940 | Grevy's zebra | DJI_0068, DJI_0069, DJI_0070, DJI_0071 |
56
+ | `session_4` | 2023-01-20 | 33,960 | Grevy's zebra | DJI_0142, DJI_0143, DJI_0144, DJI_0145, DJI_0146, DJI_0147 |
57
+ | `session_5` | 2023-01-21 | 24,106 | Giraffe, Plains and Grevy's zebras | DJI_0206, DJI_0208, DJI_0210, DJI_0211 |
58
+ | **Total Frames:** | | **104,062** | | |
59
 
60
  This table shows the data collected at Mpala Research Center in Laikipia, Kenya, with session information, dates, frame counts, and primary species observed.
61
 
 
66
  ## Dataset Structure
67
  ```​
68
  /dataset/
69
+ classes.txt
70
+ session_1/
71
+ DJI_0001/
72
+ partition_1/
73
+ DJI_0001_000000.jpg
74
+ DJI_0001_000001.txt
75
+ ...
76
+ DJI_0001_004999.txt
77
+ partition_2/
78
+ DJI_0001_005000.jpg
79
+ DJI_0001_005000.txt
80
+ ...
81
+ DJI_0001_008700.txt
82
+ DJI_0002/
83
+ DJI_0002_000000.jpg
84
+ DJI_0002_000001.txt
85
+ ...
86
+ DJI_0002_008721.txt
87
+ metadata.txt
88
+ session_2/
89
+ DJI_0005/
90
+ DJI_0005_001260.jpg
91
+ DJI_0005_001260.txt
92
+ ...
93
+ DJI_0005_008715.txt
94
+ DJI_0006/
95
+ partition_1/
96
+ DJI_0006_000000.jpg
97
+ DJI_0006_000001.txt
98
+ ...
99
+ DJI_0006_005351.txt
100
+ partition_2/
101
+ DJI_0006_005352.jpg
102
+ DJI_0006_005352.txt
103
+ ...
104
+ DJI_0006_008719.txt
105
+ metadata.txt
106
+ session_3/
107
+ DJI_0068/
108
+ DJI_0068_000780.jpg
109
+ DJI_0068_000780.txt
110
+ ...
111
+ DJI_0068_005790.txt
112
+ DJI_0069/
113
+ partition_1/
114
+ DJI_0069_000000.jpg
115
+ DJI_0069_000001.txt
116
+ ...
117
+ DJI_0069_004999.txt
118
+ partition_2/
119
+ DJI_0069_005000.jpg
120
+ DJI_0069_005000.txt
121
+ ...
122
+ DJI_0069_005815.txt
123
+ DJI_0070/
124
+ partition_1/
125
+ DJI_0070_000000.jpg
126
+ DJI_0070_000001.txt
127
+ ...
128
+ DJI_0069_004999.txt
129
+ partition_2/
130
+ DJI_0070_005000.jpg
131
+ DJI_0070_005000.txt
132
+ ...
133
+ DJI_0070_005812.txt
134
+ DJI_0071/
135
+ DJI_0071_000000.jpg
136
+ DJI_0071_000000.txt
137
+ ...
138
+ DJI_0071_001357.txt
139
+ metadata.txt
140
+ session_4/
141
+ DJI_0142/
142
+ partition_1/
143
+ DJI_0142_000000.jpg
144
+ DJI_0142_000000.txt
145
+ ...
146
+ DJI_0142_002999.txt
147
+ partition_2/
148
+ DJI_0142_003000.jpg
149
+ DJI_0142_003000.txt
150
+ ...
151
+ DJI_0142_005799.txt
152
+ DJI_0143/
153
+ partition_1/
154
+ DJI_0143_000000.jpg
155
+ DJI_0143_000000.txt
156
+ ...
157
+ DJI_0143_002999.txt
158
+ partition_2/
159
+ DJI_0143_003000.jpg
160
+ DJI_0143_003000.txt
161
+ ...
162
+ DJI_0143_005816.txt
163
+ DJI_0144/
164
+ partition_1/
165
+ DJI_0144_000000.jpg
166
+ DJI_0144_000000.txt
167
+ ...
168
+ DJI_0144_002999.txt
169
+ partition_2/
170
+ DJI_0144_003000.jpg
171
+ DJI_0144_003000.txt
172
+ ...
173
+ DJI_0144_005790.txt
174
+ DJI_0145/
175
+ partition_1/
176
+ DJI_0145_000000.jpg
177
+ DJI_0145_000000.txt
178
+ ...
179
+ DJI_0145_002999.txt
180
+ partition_2/
181
+ DJI_0145_003000.jpg
182
+ DJI_0145_003000.txt
183
+ ...
184
+ DJI_0145_005811.txt
185
+ DJI_0146/
186
+ partition_1/
187
+ DJI_0146_000000.jpg
188
+ DJI_0146_000000.txt
189
+ ...
190
+ DJI_0146_002999.txt
191
+ partition_2/
192
+ DJI_0146_003000.jpg
193
+ DJI_0146_003000.txt
194
+ ...
195
+ DJI_0146_005809.txt
196
+ DJI_0147/
197
+ partition_1/
198
+ DJI_0147_000000.jpg
199
+ DJI_0147_000000.txt
200
+ ...
201
+ DJI_0147_002999.txt
202
+ partition_2/
203
+ DJI_0147_003000.jpg
204
+ DJI_0147_003000.txt
205
+ ...
206
+ DJI_0147_005130.txt
207
+ metadata.txt
208
+ session_5/
209
+ DJI_0206/
210
+ partition_1/
211
+ DJI_0206_000000.jpg
212
+ DJI_0206_000000.txt
213
+ ...
214
+ DJI_0206_002499.txt
215
+ partition_2/
216
+ DJI_0206_002500.jpg
217
+ DJI_0206_002500.txt
218
+ ...
219
+ DJI_0206_004999.txt
220
+ partition_3/
221
+ DJI_0206_005000.jpg
222
+ DJI_0206_005000.txt
223
+ ...
224
+ DJI_0206_005802.txt
225
+ DJI_0208/
226
+ partition_1/
227
+ DJI_0208_000000.jpg
228
+ DJI_0208_000000.txt
229
+ ...
230
+ DJI_0208_002999.txt
231
+ partition_2/
232
+ DJI_0208_003000.jpg
233
+ DJI_0208_003000.txt
234
+ ...
235
+ DJI_0208_005810.txt
236
+ DJI_0210/
237
+ partition_1/
238
+ DJI_0210_000000.jpg
239
+ DJI_0210_000000.txt
240
+ ...
241
+ DJI_0210_002999.txt
242
+ partition_2/
243
+ DJI_0210_003000.jpg
244
+ DJI_0210_003000.txt
245
+ ...
246
+ DJI_0210_005811.txt
247
+ DJI_0211/
248
+ partition_1/
249
+ DJI_0211_000000.jpg
250
+ DJI_0211_000000.txt
251
+ ...
252
+ DJI_0211_002999.txt
253
+ partition_2/
254
+ DJI_0211_003000.jpg
255
+ DJI_0211_003000.txt
256
+ ...
257
+ DJI_0211_005809.txt
258
+ metadata.txt
259
+
260
  ```
261
 
262
  ### Data Instances
263
  All images are named <video_id_frame>.jpg, each within a folder named for the date of the session. The annotations are in COCO format and are stored in a corresponding .txt file with the same name as the image.
264
 
265
+ Note on data partitions: DJI saves video files into 3GB chunks, so each session is divided into multiple video files. HuggingFace limits folders to 10,000 files per folder, so each video file is further divided into partitions of 10,000 files. The partition folders are named `partition_1`, `partition_2`, etc. The original video files are not included in the dataset.
266
+
267
 
268
  ### Data Fields
269
 
270
  **classes.txt**:
271
+ - `0`: zebra
272
+ - `1`: giraffe
273
+ - `2`: onager
274
+ - `3`: dog
275
 
276
  **frame_id.txt**:
277
  - `class`: Class of the object in the image (0 for zebra)
 
417
  year={2024}
418
  }
419
  ```
420
+
421
+
422
  ```​
423
+ @article{kline2025wildwing,
424
  title={WildWing: An open-source, autonomous and affordable UAS for animal behaviour video monitoring},
425
  author={Kline, Jenna and Zhong, Alison and Irizarry, Kevyn and Stewart, Charles V and Stewart, Christopher and Rubenstein, Daniel I and Berger-Wolf, Tanya},
426
  journal={Methods in Ecology and Evolution},
 
430
  }
431
  ```
432
 
 
 
 
 
 
 
 
 
 
 
 
 
433
 
434
 
435
  ## Acknowledgements