VOT2020 Benchmark

The VOT challenges provide the tracking community with a precisely defined and repeatable way of comparing short-term trackers and long-term trackers as well as a common platform for discussing the evaluation and advancements made in the field of visual tracking. Following seven highly successful VOT challenges, the 8th Visual Object Tracking Challenge VOT2020 and workshop was held in conjunction with ECCV 2020.

Highlights

• Five challenges planned
• Toolkit re-implemented in Python
• A new performance evaluation protocol introduced for short-term tracker evaluation - more information here
• 2020 datasets refreshed in VOT-ST challenge
• Bounding boxes VOT-ST2020 & VOT-RT2020 will be replaced by segmentation masks (performance evaluated on segmentation)
• Authors of well-performing trackers will be invited to coauthor the VOT2020 results papers.
• Authors of top-performing trackers will be invited to present their trackers at VOT2020 workshop.

Challenges

• VOT short-term tracking challenge VOT-ST2020 - Robust short-term tracking under appearance variation, occlusion and clutter. Targets annotated by segmentation masks.
• VOT short-term real-time challenge VOT-RT2020 - Robust short-term tracking under time constraints. Targets annotated by segmentation masks.
• VOT long-term tracking challenge VOT-LT2020 - Robust tracking with target disappearance.
• VOT thermal and infrared short-term tracking challenge VOT-RGBT2020 - Multi-modality tracking under RGB and T partial modality de-synchronization (challenge will be opened later).
• VOT color and depth long-term tracking challenge VOT-RGBD2020 - Using depth to improve LT RGB tracking with target disappearance.

Sponsors

The VOT2020 workshop was sposored by the Faculty of Computer and Information Science, University of Ljubljana.