Optimizing rgb-d semantic segmentation through multi-modal interaction and pooling attention

TL;DR

This paper introduces MIPANet, a novel neural network that enhances RGB-D semantic segmentation by effectively fusing multi-modal data and applying pooling attention, leading to improved performance on indoor scene datasets.

Contribution

The paper proposes MIPANet with a Multi-modal Interaction Fusion Module and Pooling Attention Modules, advancing multi-modal fusion and feature enhancement in RGB-D segmentation.

Findings

Outperforms existing methods on NYUDv2 dataset

Achieves superior results on SUN-RGBD dataset

Demonstrates significant improvement in segmentation accuracy

Abstract

Semantic segmentation of RGB-D images involves understanding the appearance and spatial relationships of objects within a scene, which requires careful consideration of various factors. However, in indoor environments, the simple input of RGB and depth images often results in a relatively limited acquisition of semantic and spatial information, leading to suboptimal segmentation outcomes. To address this, we propose the Multi-modal Interaction and Pooling Attention Network (MIPANet), a novel approach designed to harness the interactive synergy between RGB and depth modalities, optimizing the utilization of complementary information. Specifically, we incorporate a Multi-modal Interaction Fusion Module (MIM) into the deepest layers of the network. This module is engineered to facilitate the fusion of RGB and depth information, allowing for mutual enhancement and correction. Additionally, we introduce a Pooling Attention Module (PAM) at various stages of the encoder. This module serves to amplify the features extracted by the network and integrates the module's output into the decoder in a targeted manner, significantly improving semantic segmentation performance. Our experimental results demonstrate that MIPANet outperforms existing methods on two indoor scene datasets, NYUDv2 and SUN-RGBD, underscoring its effectiveness in enhancing RGB-D semantic segmentation.