CoTracker3: simpler and better point tracking by pseudo-labeling real videos

Conference item

We introduce CoTracker3, a new state-of-the-art point tracker. With CoTracker3, we revisit the design of recent trackers, removing components and reducing the number of parameters while also improving performance. We also explore the interplay of synthetic and real data. Recent trackers are trained on synthetic videos due to the difficulty of collecting tracking annotations for real data. However, this can result in suboptimal performance due to the statistical gap between synthetic and real videos. We thus suggest using off-the-shelf trackers as teachers to annotate real videos with pseudo-labels. Compared to other recent attempts at using real data for learning trackers, this scheme is much simpler and achieves better results using 1,000 times less data. CoTracker3 is available here in online (causal) and offline variants.

Authors: Karaev, N; Makarov, I; Wang, J; Neverova, N; Vedaldi, A

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: IEEE
• Host title: 2025 IEEE/CVF International Conference on Computer Vision (ICCV)
• Pages: 1-10
• Publication date: 2026-04-29
• Acceptance date: 2025-07-23
• Event title: International Conference on Computer Vision (ICCV 2025)
• Event location: Honolulu, Hawai'i, USA
• Event website: https://iccv.thecvf.com/
• Event start date: 2025-10-19
• Event end date: 2025-10-23
• DOI: 10.1109/ICCV51701.2025.00568
• EISSN: 2380-7504
• ISSN: 1550-5499
• EISBN: 9798331587758
• ISBN: 9798331587765
• Language: English
• Keywords: feeds; filtering; filters; flexible printed circuits; protocols; videos; pixel; communication systems; lora; video equipment