Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,24 @@
|
|
| 1 |
-
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# PanoMouse Dataset
|
| 2 |
+
|
| 3 |
+
PanoMouse is the official dataset for the paper:
|
| 4 |
+
|
| 5 |
+
> **Mouse Total Capture: A Multi-view Dataset for 3D Motion and Expression
|
| 6 |
+
> Capture of Freely Moving Mouse**
|
| 7 |
+
> Anonymous Authors, NeurIPS 2026
|
| 8 |
+
|
| 9 |
+
PanoMouse supports simultaneous whole-body motion and facial expression capture
|
| 10 |
+
of freely moving mice via a 24-camera multiview system.
|
| 11 |
+
|
| 12 |
+
## Dataset Overview
|
| 13 |
+
|
| 14 |
+
- **Cameras**: 24 cameras distributed across top-middle-bottom 3 layers, 360° coverage
|
| 15 |
+
- **Frames**: Over 3.4 million frames
|
| 16 |
+
- **Annotations**: 7,293 frames annotated with **92 whole-body keypoints**
|
| 17 |
+
- **Tasks**: 2D pose estimation, 3D pose estimation, behavior classification
|
| 18 |
+
|
| 19 |
+
## Repository Structure
|
| 20 |
+
PanoMouse/
|
| 21 |
+
├── action_annotations/ # Action category labels for behavior classification
|
| 22 |
+
├── annotations/ # 2D and 3D pose annotations (keypoint coordinates)
|
| 23 |
+
├── videos-1K.zip # Multi-view synchronized videos (1K subset)
|
| 24 |
+
└── images.zip # Multi-view image frames
|