RF-Net: An End-to-End Image Matching Network based on Receptive Field

Abstract

This paper proposes a new end-to-end trainable matching network based on receptive field, RF-Net, to compute sparse correspondence between images. Building end-to-end trainable matching framework is desirable and challenging. The very recent approach, LF-Net, successfully embeds the entire feature extraction pipeline into a jointly trainable pipeline, and produces the state-of-the-art matching results. This paper introduces two modifications to the structure of LF-Net. First, we propose to construct receptive feature maps, which lead to more effective keypoint detection. Second, we introduce a general loss function term, neighbor mask, to facilitate training patch selection. This results in improved stability in descriptor training. We trained RF-Net on the open dataset HPatches, and compared it with other methods on multiple benchmark datasets. Experiments show that RF-Net outperforms existing state-of-the-art methods.

arXiv comment9 pages, 6 figures

References in corpus (6)

• Adam: A Method for Stochastic Optimization
• Instance Normalization: The Missing Ingredient for Fast Stylization
• Faster and better: a machine learning approach to corner detection
• Action Recognition with Trajectory-Pooled Deep-Convolutional Descriptors
• TILDE: A Temporally Invariant Learned DEtector
• Working hard to know your neighbor's margins: Local descriptor learning loss