# DMSA: Dynamic Multi-scale Unsupervised Semantic Segmentation Based on Adaptive Affinity

## TL;DR

This paper introduces DMSA, an unsupervised semantic segmentation framework that leverages multi-scale feature extraction, adaptive dilation, and pixel refinement to improve accuracy and convergence speed on standard datasets.

## Contribution

The paper presents a novel end-to-end unsupervised segmentation architecture combining ASPP, dynamic dilation, and pixel-adaptive refinement modules for enhanced performance.

## Key findings

- Improved MIoU by 2.0 on COCO 80 dataset
- Enhanced accuracy by 5.39 on COCO 80 dataset
- Faster convergence with the PAR module

## Abstract

The proposed method in this paper proposes an end-to-end unsupervised semantic segmentation architecture DMSA based on four loss functions. The framework uses Atrous Spatial Pyramid Pooling (ASPP) module to enhance feature extraction. At the same time, a dynamic dilation strategy is designed to better capture multi-scale context information. Secondly, a Pixel-Adaptive Refinement (PAR) module is introduced, which can adaptively refine the initial pseudo labels after feature fusion to obtain high quality pseudo labels. Experiments show that the proposed DSMA framework is superior to the existing methods on the saliency dataset. On the COCO 80 dataset, the MIoU is improved by 2.0, and the accuracy is improved by 5.39. On the Pascal VOC 2012 Augmented dataset, the MIoU is improved by 4.9, and the accuracy is improved by 3.4. In addition, the convergence speed of the model is also greatly improved after the introduction of the PAR module.