OVTR: End-to-End Open-Vocabulary Multiple Object Tracking with Transformer
Keywords: Multiple-Object Tracking, Transformer, End-to-End, Open-Vocabulary
Abstract: Open-vocabulary multiple object tracking aims to generalize trackers to unseen categories during training, enabling their application across a variety of real-world scenarios. However, the existing open-vocabulary tracker, which relies on off-the-shelf open-vocabulary detector, perceives categories and locations independently in each frame, causing instability and making it vulnerable to similar appearances and irregular motion in diverse scenes. In this paper, we propose OVTR (End-to-End Open-Vocabulary Multiple Object Tracking with TRansformer), the first end-to-end open-vocabulary tracker that models motion, appearance, and category simultaneously. To achieve stable classification and continuous tracking, we designed the CIP (Category Information Propagation) strategy, which establishes multiple high-level category information priors for subsequent frames. Additionally, we introduce a dual-branch structure for generalization capability and deep multimodal interaction, and incorporate protective strategies in the decoder to enhance performance. Notably, our method does not require proposals that contain novel categories, yet still achieves strong results on the open-vocabulary MOT benchmark. Moreover, experiment transferring the model to other dataset demonstrates its effective adaptability.
Primary Area: applications to computer vision, audio, language, and other modalities
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 2059
Loading