Dataset Versions
Versions
2024-06-05 6:52am
v23
· 7 months ago
2024-05-26 8:18am
v22
· 7 months ago
2024-05-24 10:35am
v21
· 7 months ago
2024-05-22 2:12pm
v20
· 7 months ago
2024-05-10 11:29pm
v19
· 8 months ago
2024-05-07 12:56pm
v17
· 8 months ago
2024-05-02 8:19am
v16
· 8 months ago
2024-04-20 4:03pm
v15
· 8 months ago
2024-04-13 3:11pm
v14
· 9 months ago
2024-03-26 9:54pm
v13
· 9 months ago
2024-03-23 10:26pm
v12
· 9 months ago
2024-03-21 11:19pm
v11
· 9 months ago
2024-03-21 11:11pm
v10
· 9 months ago
2024-03-21 12:50am
v9
· 9 months ago
2024-03-20 10:44pm
v4
· 9 months ago
2024-03-19 3:23pm
v3
· 9 months ago
2024-03-07 2:38pm
v2
· 10 months ago
2024-01-23 7:48pm
v1
· a year ago
v17
2024-05-07 12:56pm
Generated on May 7, 2024
Popular Download Formats
YOLOv11
TXT annotations and YAML config used with YOLOv11.
YOLOv9
TXT annotations and YAML config used with YOLOv9.
YOLOv8
TXT annotations and YAML config used with YOLOv8.
YOLOv5
TXT annotations and YAML config used with YOLOv5.
YOLOv7
TXT annotations and YAML config used with YOLOv7.
COCO JSON
COCO JSON annotations are used with EfficientDet Pytorch and Detectron 2.
YOLO Darknet
Darknet TXT annotations used with YOLO Darknet (both v3 and v4) and YOLOv3 PyTorch.
Pascal VOC XML
Common XML annotation format for local data munging (pioneered by ImageNet).
TFRecord
TFRecord binary format used for both Tensorflow 1.5 and Tensorflow 2.0 Object Detection models.
PaliGemma
PaliGemma JSONL format used for fine-tuning PaliGemma, Google's open multimodal vision model.
CreateML JSON
CreateML JSON format is used with Apple's CreateML and Turi Create tools.
Other Formats
Choose another format.
5836 Total Images
View All Images
Dataset Split
Train Set 85%
4941Images
Valid Set 12%
706Images
Test Set 3%
189Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Augmentations
Outputs per training example: 3
Rotation: Between -16° and +16°
Exposure: Between -10% and +10%
Noise: Up to 0.1% of pixels
Bounding Box: Shear: ±5° Horizontal, ±5° Vertical
Bounding Box: Exposure: Between -10% and +10%
Bounding Box: Noise: Up to 0.02% of pixels