mvHOTA: A multi-view higher order tracking accuracy metric to measure spatial and temporal associations in multi-point detection.
Lalith SharanHalvar KelmGabriele RomanoMatthias KarckRaffaele De SimoneSandy EngelhardtPublished in: CoRR (2022)
Keyphrases
- multi view
- spatial and temporal
- higher order
- tracking accuracy
- spatio temporal
- multiple views
- single view
- detection rate
- d objects
- camera calibration
- depth map
- space time
- three dimensional
- semi supervised
- dynamic scenes
- multiple cameras
- object tracking
- video frames
- object detection
- target tracking
- detection algorithm
- camera parameters
- detection method
- particle filter
- partial occlusion
- similarity measure
- face tracking
- machine learning