OpenMMLab Video Perception Toolbox. It supports Video Object Detection (VID), Multiple Object Tracking (MOT), Single Object Tracking (SOT), Video Instance Segmentation (VIS) with a unified framework.
Can the existing project implement mask function for multi-Object tracking (instead of simple box selection of target objects).If not, how can I modify the existing code
mikel-brostrom
commented
2 days ago
Can the existing project implement mask function for multi-Object tracking (instead of simple box selection of target objects).If not, how can I modify the existing code