OmniTrack++: Omnidirectional Multi-Object Tracking by Learning Large-FoV Trajectory Feedback
Researchers have introduced OmniTrack++, an advanced framework for omnidirectional multi-object tracking designed to overcome challenges like panoramic distortion and identity ambiguity. The system utilizes a feedback-driven approach, incorporating a DynamicSSM block for feature stabilization and FlexiTrack Instances for precise localization and association. To enhance long-term tracking, an ExpertTrack Memory consolidates appearance cues, while a Tracklet Management module adaptively switches between tracking modes based on scene dynamics. The team also released the EmboTrack benchmark, featuring new datasets like QuadTrack and BipTrack, to facilitate evaluation in real-world panoramic scenarios. AI
IMPACT Introduces a new benchmark and tracking method that could improve perception systems in robotics and autonomous systems.