Contrastive Multi-Task Dense Prediction
Siwei Yang, Hanrong Ye, Dan Xu
Abstract
This paper targets the problem of multi-task dense prediction which aims to achieve simultaneous learning and inference on a bunch of multiple dense prediction tasks in a single framework. A core objective in design is how to effectively model cross-task interactions to achieve a comprehensive improvement on different tasks based on their inherent complementarity and consistency. Existing works typically design extra expensive distillation modules to perform explicit interaction computations among different task-specific features in both training and inference, bringing difficulty in adaptation for different task sets, and reducing efficiency due to clearly increased size of multi-task models. In contrast, we introduce feature-wise contrastive consistency into modeling the cross-task interactions for multi-task dense prediction. We propose a novel multi-task contrastive regularization method based on the consistency to effectively boost the representation learning of the different sub-tasks, which can also be easily generalized to different multi-task dense prediction frameworks, and costs no additional computation in the inference. Extensive experiments on two challenging datasets (i.e. NYUD-v2 and Pascal-Context) clearly demonstrate the superiority of the proposed multi-task contrastive learning approach for dense predictions, establishing new state-of-the-art performances.
BibTeX
@article{Yang_Ye_Xu_2023, title={Contrastive Multi-Task Dense Prediction}, volume={37}, url={https://ojs.aaai.org/index.php/AAAI/article/view/25424}, DOI={10.1609/aaai.v37i3.25424}, abstractNote={This paper targets the problem of multi-task dense prediction
which aims to achieve simultaneous learning and inference on
a bunch of multiple dense prediction tasks in a single framework. A core objective in design is how to effectively model
cross-task interactions to achieve a comprehensive improvement on different tasks based on their inherent complementarity and consistency. Existing works typically design extra
expensive distillation modules to perform explicit interaction
computations among different task-specific features in both
training and inference, bringing difficulty in adaptation for
different task sets, and reducing efficiency due to clearly increased size of multi-task models. In contrast, we introduce
feature-wise contrastive consistency into modeling the cross-task interactions for multi-task dense prediction. We propose
a novel multi-task contrastive regularization method based on
the consistency to effectively boost the representation learning of the different sub-tasks, which can also be easily generalized to different multi-task dense prediction frameworks,
and costs no additional computation in the inference. Extensive experiments on two challenging datasets (i.e. NYUD-v2
and Pascal-Context) clearly demonstrate the superiority of the
proposed multi-task contrastive learning approach for dense
predictions, establishing new state-of-the-art performances.}, number={3}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Yang, Siwei and Ye, Hanrong and Xu, Dan}, year={2023}, month={Jun.}, pages={3190-3197} }