CVApr 6, 2023

Learning Neural Eigenfunctions for Unsupervised Semantic Segmentation

arXiv:2304.02841v17 citationsh-index: 16
Originality Incremental advance
AI Analysis

This work addresses a long-standing challenge in computer vision for unsupervised semantic segmentation, offering an incremental improvement by making spectral clustering more efficient and flexible.

The paper tackles the inefficiency and inflexibility of spectral clustering for unsupervised semantic segmentation by introducing neural eigenfunctions that directly output discrete clustering assignments, achieving significant performance gains on Pascal Context, Cityscapes, and ADE20K benchmarks.

Unsupervised semantic segmentation is a long-standing challenge in computer vision with great significance. Spectral clustering is a theoretically grounded solution to it where the spectral embeddings for pixels are computed to construct distinct clusters. Despite recent progress in enhancing spectral clustering with powerful pre-trained models, current approaches still suffer from inefficiencies in spectral decomposition and inflexibility in applying them to the test data. This work addresses these issues by casting spectral clustering as a parametric approach that employs neural network-based eigenfunctions to produce spectral embeddings. The outputs of the neural eigenfunctions are further restricted to discrete vectors that indicate clustering assignments directly. As a result, an end-to-end NN-based paradigm of spectral clustering emerges. In practice, the neural eigenfunctions are lightweight and take the features from pre-trained models as inputs, improving training efficiency and unleashing the potential of pre-trained models for dense prediction. We conduct extensive empirical studies to validate the effectiveness of our approach and observe significant performance gains over competitive baselines on Pascal Context, Cityscapes, and ADE20K benchmarks.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes