Researchers from Bytedance and Dalian University Propose 🦄 ‘Unicorn’: a Unified Computer Vision Approach to Address Four Tracking Tasks Using a Single Model with the Same Model Parameters

Object tracking is one of the core applications in the field of computer vision. It constructs pixel-level or instance-level connections amongst frames and produces trajectories in the form of boxes (masks). There are four primary categories of object tracking that include: Single Object Tracking (SOT); Video Object Segmentation (VOS); Multiple Object Tracking (MOT); Multi-Object Tracking and Segmentation (MOTS). Most of the applications solve only one category of tracking. A unified model is required to handle all the categories of tracking …

bytedance computer computer vision computervision researchers tracking unicorn university vision