Demystifying Unsupervised Semantic Correspondence Estimation
This addresses the problem of semantic correspondence estimation without labeled data for computer vision researchers, though it appears incremental.
The paper tackles unsupervised semantic correspondence estimation by evaluating existing methods, developing a diagnostic framework with a new metric, and introducing an approach that leverages pre-trained features to improve matching. Their method achieves significantly better performance than current state-of-the-art methods.
We explore semantic correspondence estimation through the lens of unsupervised learning. We thoroughly evaluate several recently proposed unsupervised methods across multiple challenging datasets using a standardized evaluation protocol where we vary factors such as the backbone architecture, the pre-training strategy, and the pre-training and finetuning datasets. To better understand the failure modes of these methods, and in order to provide a clearer path for improvement, we provide a new diagnostic framework along with a new performance metric that is better suited to the semantic matching task. Finally, we introduce a new unsupervised correspondence approach which utilizes the strength of pre-trained features while encouraging better matches during training. This results in significantly better matching performance compared to current state-of-the-art methods.