Synthetic multiview video sequences and 3D annotations for cattle detection and action recognition. The data includes synchronized camera angles and behavioral labels for livestock monitoring in simulated environments.
Use Cases
- Train 3D object detection models using the provided spatial coordinates and multiview frames
- Develop action recognition systems to identify cattle behaviors across temporal sequences
- Benchmark multiview fusion algorithms to resolve occlusions in livestock monitoring
Strengths
- Synthetic multiview video frames for synchronized 3D spatial analysis
- Ground truth 3D bounding boxes for individual cattle detection
- Temporal action labels for classifying specific livestock behaviors
- Integrated toolkits for handling multiview synthetic data formats