Skip to main content

Tunable Hybrid Proposal Networks for the Open World

Publication ,  Conference
Inkawhich, M; Inkawhich, N; Li, H; Chen, Y
Published in: Proceedings - 2024 IEEE Winter Conference on Applications of Computer Vision, WACV 2024
January 3, 2024

Current state-of-the-art object proposal networks are trained with a closed-world assumption, meaning they learn to only detect objects of the training classes. These models fail to provide high recall in open-world environments where important novel objects may be encountered. While a handful of recent works attempt to tackle this problem, they fail to consider that the optimal behavior of a proposal network can vary significantly depending on the data and application. Our goal is to provide a flexible proposal solution that can be easily tuned to suit a variety of open-world settings. To this end, we design a Tunable Hybrid Proposal Network (THPN) that leverages an adjustable hybrid architecture, a novel self-training procedure, and dynamic loss components to optimize the tradeoff between known and unknown object detection performance. To thoroughly evaluate our method, we devise several new challenges which invoke varying degrees of label bias by altering known class diversity and label count. We find that in every task, THPN easily outperforms existing baselines (e.g., RPN, OLN). Our method is also highly data efficient, surpassing baseline recall with a fraction of the labeled data.

Duke Scholars

Published In

Proceedings - 2024 IEEE Winter Conference on Applications of Computer Vision, WACV 2024

DOI

Publication Date

January 3, 2024

Start / End Page

1977 / 1988
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Inkawhich, M., Inkawhich, N., Li, H., & Chen, Y. (2024). Tunable Hybrid Proposal Networks for the Open World. In Proceedings - 2024 IEEE Winter Conference on Applications of Computer Vision, WACV 2024 (pp. 1977–1988). https://doi.org/10.1109/WACV57701.2024.00199
Inkawhich, M., N. Inkawhich, H. Li, and Y. Chen. “Tunable Hybrid Proposal Networks for the Open World.” In Proceedings - 2024 IEEE Winter Conference on Applications of Computer Vision, WACV 2024, 1977–88, 2024. https://doi.org/10.1109/WACV57701.2024.00199.
Inkawhich M, Inkawhich N, Li H, Chen Y. Tunable Hybrid Proposal Networks for the Open World. In: Proceedings - 2024 IEEE Winter Conference on Applications of Computer Vision, WACV 2024. 2024. p. 1977–88.
Inkawhich, M., et al. “Tunable Hybrid Proposal Networks for the Open World.” Proceedings - 2024 IEEE Winter Conference on Applications of Computer Vision, WACV 2024, 2024, pp. 1977–88. Scopus, doi:10.1109/WACV57701.2024.00199.
Inkawhich M, Inkawhich N, Li H, Chen Y. Tunable Hybrid Proposal Networks for the Open World. Proceedings - 2024 IEEE Winter Conference on Applications of Computer Vision, WACV 2024. 2024. p. 1977–1988.

Published In

Proceedings - 2024 IEEE Winter Conference on Applications of Computer Vision, WACV 2024

DOI

Publication Date

January 3, 2024

Start / End Page

1977 / 1988