Publication Type
Conference Proceeding Article
Version
publishedVersion
Publication Date
5-2023
Abstract
Prompt tuning with large-scale pretrained vision-language models empowers open-vocabulary prediction trained on limited base categories, e.g., object classification and detection. In this paper, we propose compositional prompt tuning with motion cues: an extended prompt tuning paradigm for compositional predictions of video data. In particular, we present Relation Prompt (RePro) for Open-vocabulary Video Visual Relation Detection (Open-VidVRD), where conventional prompt tuning is easily biased to certain subject-object combinations and motion patterns. To this end, RePro addresses the two technical challenges of Open-VidVRD: 1) the prompt tokens should respect the two different semantic roles of subject and object, and 2) the tuning should account for the diverse spatiotemporal motion patterns of the subject-object compositions. Our RePro achieves a new state-of-the-art performance on two VidVRD benchmarks of not only the base training object and predicate categories, but also the unseen ones. Extensive ablations also demonstrate the effectiveness of the proposed compositional and multi-mode design of prompt.
Discipline
Databases and Information Systems | Graphics and Human Computer Interfaces
Research Areas
Data Science and Engineering
Publication
Proceedings of the eleventh International Conference on Learning Representations, Kigali, Rwanda, 2023 May 1-5
First Page
1
Last Page
16
Publisher
ICLR
City or Country
Kigali, Rwanda
Citation
GAO, Kaifeng; CHEN, Long; ZHANG, Hanwang; XIAO, Jun; and SUN, Qianru.
Compositional prompt tuning with motion cues for open-vocabulary video relation detection. (2023). Proceedings of the eleventh International Conference on Learning Representations, Kigali, Rwanda, 2023 May 1-5. 1-16.
Available at: https://ink.library.smu.edu.sg/sis_research/8102
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Included in
Databases and Information Systems Commons, Graphics and Human Computer Interfaces Commons