MaskRIS: Semantic Distortion-aware Data Augmentation for Referring Image Segmentation

TL;DR

MaskRIS introduces a novel data augmentation framework for Referring Image Segmentation that employs image and text masking combined with distortion-aware learning, significantly improving robustness and achieving state-of-the-art results.

Contribution

The paper proposes MaskRIS, a new training framework utilizing masking strategies and distortion-aware learning to enhance RIS performance beyond existing methods.

Findings

Outperforms existing RIS methods on multiple datasets

Improves robustness to occlusions and linguistic complexities

Effective in both supervised and weakly supervised settings

Abstract

Referring Image Segmentation (RIS) is an advanced vision-language task that involves identifying and segmenting objects within an image as described by free-form text descriptions. While previous studies focused on aligning visual and language features, exploring training techniques, such as data augmentation, remains underexplored. In this work, we explore effective data augmentation for RIS and propose a novel training framework called Masked Referring Image Segmentation (MaskRIS). We observe that the conventional image augmentations fall short of RIS, leading to performance degradation, while simple random masking significantly enhances the performance of RIS. MaskRIS uses both image and text masking, followed by Distortion-aware Contextual Learning (DCL) to fully exploit the benefits of the masking strategy. This approach can improve the model's robustness to occlusions, incomplete information, and various linguistic complexities, resulting in a significant performance improvement. Experiments demonstrate that MaskRIS can easily be applied to various RIS models, outperforming existing methods in both fully supervised and weakly supervised settings. Finally, MaskRIS achieves new state-of-the-art performance on RefCOCO, RefCOCO+, and RefCOCOg datasets. Code is available at https://github.com/naver-ai/maskris.