DynaSeg: A Deep Dynamic Fusion Method for Unsupervised Image Segmentation Incorporating Feature Similarity and Spatial Continuity

arXiv:2405.05477v1 Announce Type: new
Abstract: Our work tackles the fundamental challenge of image segmentation in computer vision, which is crucial for diverse applications. While supervised methods demonstrate proficiency, their reliance on extensive pixel-level annotations limits scalability. In response to this challenge, we present an enhanced unsupervised Convolutional Neural Network (CNN)-based algorithm called DynaSeg. Unlike traditional approaches that rely on a fixed weight factor to balance feature similarity and spatial continuity, requiring manual adjustments, our novel, dynamic weighting scheme automates parameter …

abstract annotations applications arxiv challenge computer computer vision continuity cs.cv diverse diverse applications dynamic feature fundamental fusion image pixel reliance scalability segmentation spatial type unsupervised vision while work