Learning Modality-agnostic Representation for Semantic Segmentation from Any Modalities

TL;DR

This paper introduces Any2Seg, a framework that learns modality-agnostic representations for semantic segmentation, enabling robust performance across various modalities and conditions by leveraging knowledge distillation from vision-language models.

Contribution

The paper proposes a novel framework combining language-guided semantic correlation distillation and modality-agnostic feature fusion to improve multi-modal semantic segmentation robustness.

Findings

Achieves state-of-the-art results on benchmarks with four modalities.

Significantly improves performance in modality-incomplete scenarios.

Demonstrates robustness across diverse visual conditions.

Abstract

Image modality is not perfect as it often fails in certain conditions, e.g., night and fast motion. This significantly limits the robustness and versatility of existing multi-modal (i.e., Image+X) semantic segmentation methods when confronting modality absence or failure, as often occurred in real-world applications. Inspired by the open-world learning capability of multi-modal vision-language models (MVLMs), we explore a new direction in learning the modality-agnostic representation via knowledge distillation (KD) from MVLMs. Intuitively, we propose Any2Seg, a novel framework that can achieve robust segmentation from any combination of modalities in any visual conditions. Specifically, we first introduce a novel language-guided semantic correlation distillation (LSCD) module to transfer both inter-modal and intra-modal semantic knowledge in the embedding space from MVLMs, e.g., LanguageBind. This enables us to minimize the modality gap and alleviate semantic ambiguity to combine any modalities in any visual conditions. Then, we introduce a modality-agnostic feature fusion (MFF) module that reweights the multi-modal features based on the inter-modal correlation and selects the fine-grained feature. This way, our Any2Seg finally yields an optimal modality-agnostic representation. Extensive experiments on two benchmarks with four modalities demonstrate that Any2Seg achieves the state-of-the-art under the multi-modal setting (+3.54 mIoU) and excels in the challenging modality-incomplete setting(+19.79 mIoU).