OMH: Structured Sparsity via Optimally Matched Hierarchy for Unsupervised Semantic Segmentation
This addresses the challenge of segmenting images without labels for computer vision applications, representing an incremental advance in unsupervised learning techniques.
The paper tackles the problem of unsupervised semantic segmentation by introducing Optimally Matched Hierarchy (OMH), which imposes structured sparsity on features to encode multi-granularity information, resulting in improved segmentation performance over existing methods.
Unsupervised Semantic Segmentation (USS) involves segmenting images without relying on predefined labels, aiming to alleviate the burden of extensive human labeling. Existing methods utilize features generated by self-supervised models and specific priors for clustering. However, their clustering objectives are not involved in the optimization of the features during training. Additionally, due to the lack of clear class definitions in USS, the resulting segments may not align well with the clustering objective. In this paper, we introduce a novel approach called Optimally Matched Hierarchy (OMH) to simultaneously address the above issues. The core of our method lies in imposing structured sparsity on the feature space, which allows the features to encode information with different levels of granularity. The structure of this sparsity stems from our hierarchy (OMH). To achieve this, we learn a soft but sparse hierarchy among parallel clusters through Optimal Transport. Our OMH yields better unsupervised segmentation performance compared to existing USS methods. Our extensive experiments demonstrate the benefits of OMH when utilizing our differentiable paradigm. We will make our code publicly available.