Contrastive Multi-Task Dense Prediction

TL;DR

This paper introduces a contrastive regularization method for multi-task dense prediction that improves task performance without extra inference costs, demonstrating state-of-the-art results on benchmark datasets.

Contribution

It proposes a novel contrastive regularization technique that models cross-task interactions efficiently, enhancing multi-task learning without additional inference complexity.

Findings

Achieves superior performance on NYUD-v2 and Pascal-Context datasets.

No extra computation during inference, maintaining efficiency.

Establishes new state-of-the-art results for multi-task dense prediction.

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.