DOEI: Dual Optimization of Embedding Information for Attention-Enhanced Class Activation Maps

TL;DR

DOEI introduces a novel dual optimization approach that enhances embedding representations in attention-based weakly supervised semantic segmentation, significantly improving CAM quality and segmentation accuracy.

Contribution

The paper proposes DOEI, a new method that reconstructs embedding representations with semantic-aware attention to better align activation responses with semantic information.

Findings

Improved CAM quality and segmentation accuracy on PASCAL VOC and MS COCO datasets.

DOEI enhances the propagation and decoupling of target features in high-level semantic space.

Significant performance gains demonstrate DOEI's effectiveness as a plug-and-play module.

Abstract

Weakly supervised semantic segmentation (WSSS) typically utilizes limited semantic annotations to obtain initial Class Activation Maps (CAMs). However, due to the inadequate coupling between class activation responses and semantic information in high-dimensional space, the CAM is prone to object co-occurrence or under-activation, resulting in inferior recognition accuracy. To tackle this issue, we propose DOEI, Dual Optimization of Embedding Information, a novel approach that reconstructs embedding representations through semantic-aware attention weight matrices to optimize the expression capability of embedding information. Specifically, DOEI amplifies tokens with high confidence and suppresses those with low confidence during the class-to-patch interaction. This alignment of activation responses with semantic information strengthens the propagation and decoupling of target features, enabling the generated embeddings to more accurately represent target features in high-level semantic space. In addition, we propose a hybrid-feature alignment module in DOEI that combines RGB values, embedding-guided features, and self-attention weights to increase the reliability of candidate tokens. Comprehensive experiments show that DOEI is an effective plug-and-play module that empowers state-of-the-art visual transformer-based WSSS models to significantly improve the quality of CAMs and segmentation performance on popular benchmarks, including PASCAL VOC (+3.6%, +1.5%, +1.2% mIoU) and MS COCO (+1.2%, +1.6% mIoU). Code will be available at https://github.com/AIGeeksGroup/DOEI.