ETATrack combines visible light, heat and selected earlier frames to track objects
AI-generated summary Researchers from the Changchun Institute of Optics, Fine Mechanics and Physics of the Chinese Academy of Sciences developed ETATrack, a single-stream network that combines visible-light and thermal-infrared images with selected historical frames. On the LasHeR benchmark, the tracker achieved a success rate of 58.2% while maintaining real-time inference.