ViT-P3DE∗: Vision Transformer Based Multi-Camera Instance Association with Pseudo 3D Position Embeddings.
Minseok SeoHyuk-Jae LeeXuan Truong NguyenPublished in: IJCAI (2023)
Keyphrases
- multi camera
- pan tilt
- multi object tracking
- multiple cameras
- camera network
- object tracking
- vision system
- computer vision
- camera views
- surveillance system
- consistent labeling
- omni directional
- real time
- multi sensor
- camera placement
- multi camera systems
- dimensionality reduction
- image processing
- field of view
- pose estimation
- low dimensional
- object detection
- viewpoint