Rethinking Superpixel Segmentation from Biologically Inspired Mechanisms

TL;DR

This paper introduces a biologically inspired neural network architecture for superpixel segmentation that improves boundary adherence and semantic information by mimicking visual cortex mechanisms.

Contribution

It proposes a novel network with an Enhanced Screening Module and Boundary-Aware Label, inspired by neural structures, to enhance superpixel segmentation quality.

Findings

Improved boundary adherence in superpixels.

Enhanced semantic information representation.

Effective on BSDS500 and NYUv2 datasets.

Abstract

Recently, advancements in deep learning-based superpixel segmentation methods have brought about improvements in both the efficiency and the performance of segmentation. However, a significant challenge remains in generating superpixels that strictly adhere to object boundaries while conveying rich visual significance, especially when cross-surface color correlations may interfere with objects. Drawing inspiration from neural structure and visual mechanisms, we propose a biological network architecture comprising an Enhanced Screening Module (ESM) and a novel Boundary-Aware Label (BAL) for superpixel segmentation. The ESM enhances semantic information by simulating the interactive projection mechanisms of the visual cortex. Additionally, the BAL emulates the spatial frequency characteristics of visual cortical cells to facilitate the generation of superpixels with strong boundary adherence. We demonstrate the effectiveness of our approach through evaluations on both the BSDS500 dataset and the NYUv2 dataset.