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

TL;DR

DynaSeg is an unsupervised image segmentation method that dynamically balances feature similarity and spatial continuity, achieving state-of-the-art results without extensive hyperparameter tuning.

Contribution

It introduces a novel dynamic weighting scheme and Silhouette Score Phase for improved unsupervised segmentation, leveraging CNN features for efficiency.

Findings

Achieves 12.2% and 14.12% mIOU improvements on COCO datasets.

Demonstrates robustness across five benchmark datasets.

Provides qualitative and quantitative validation of effectiveness.

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. We introduce DynaSeg, an innovative unsupervised image segmentation approach that overcomes the challenge of balancing feature similarity and spatial continuity without relying on extensive hyperparameter tuning. Unlike traditional methods, DynaSeg employs a dynamic weighting scheme that automates parameter tuning, adapts flexibly to image characteristics, and facilitates easy integration with other segmentation networks. By incorporating a Silhouette Score Phase, DynaSeg prevents undersegmentation failures where the number of predicted clusters might converge to one. DynaSeg uses CNN-based and pre-trained ResNet feature extraction, making it computationally efficient and more straightforward than other complex models. Experimental results showcase state-of-the-art performance, achieving a 12.2% and 14.12% mIOU improvement over current unsupervised segmentation approaches on COCO-All and COCO-Stuff datasets, respectively. We provide qualitative and quantitative results on five benchmark datasets, demonstrating the efficacy of the proposed approach. Code available at Code available at https://github.com/RyersonMultimediaLab/DynaSeg