SCNet: Learning Semantic Correspondence

TL;DR

This paper introduces SCNet, a convolutional neural network that learns semantic correspondences between images by incorporating geometric consistency, outperforming previous methods on standard benchmarks.

Contribution

SCNet is a novel CNN architecture that explicitly models geometric consistency for semantic correspondence, using region proposals as matching primitives.

Findings

SCNet outperforms recent deep learning architectures.

SCNet surpasses previous hand-crafted feature methods.

The approach achieves superior results on standard benchmarks.

Abstract

This paper addresses the problem of establishing semantic correspondences between images depicting different instances of the same object or scene category. Previous approaches focus on either combining a spatial regularizer with hand-crafted features, or learning a correspondence model for appearance only. We propose instead a convolutional neural network architecture, called SCNet, for learning a geometrically plausible model for semantic correspondence. SCNet uses region proposals as matching primitives, and explicitly incorporates geometric consistency in its loss function. It is trained on image pairs obtained from the PASCAL VOC 2007 keypoint dataset, and a comparative evaluation on several standard benchmarks demonstrates that the proposed approach substantially outperforms both recent deep learning architectures and previous methods based on hand-crafted features.