Convolutional neural network architecture for geometric matching

2017/03/16 by Ignacio Rocco, Rocco, Ignacio, Relja Arandjelović +4 · 10 citations
Computer Science · Engineering · #Advanced Image and Video Retrieval Techniques #Advanced Neural Network Applications #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Robotics and Sensor-Based Localization #cs.CV #cs.LG

paper · pdf · doi:10.48550/arxiv.1703.05593

In 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017)

openalex publication_date 2017/03/16 · arxiv created 2017/04/13 · arxiv updated 2017/04/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/29

Abstract

We address the problem of determining correspondences between two images in agreement with a geometric model such as an affine or thin-plate spline transformation, and estimating its parameters. The contributions of this work are three-fold. First, we propose a convolutional neural network architecture for geometric matching. The architecture is based on three main components that mimic the standard steps of feature extraction, matching and simultaneous inlier detection and model parameter estimation, while being trainable end-to-end. Second, we demonstrate that the network parameters can be trained from synthetically generated imagery without the need for manual annotation and that our matching layer significantly increases generalization capabilities to never seen before images. Finally, we show that the same model can perform both instance-level and category-level matching giving state-of-the-art results on the challenging Proposal Flow dataset.

Citations

• Random sample consensus
• SIFT Flow: Dense Correspondence across Scenes and Its Applications
• Distinctive Image Features from Scale-Invariant Keypoints
• Deep Image Homography Estimation
• Learning to Compare Image Patches via Convolutional Neural Networks

Cited by

• Emergent Outlier View Rejection in Visual Geometry Grounded Transformers
• View-Consistent Diffusion Representations for 3D-Consistent Video Generation
• Dense Semantic Matching with VGGT Prior
• Lost in Translation? Vocabulary Alignment for Source-Free Adaptation in Open-Vocabulary Semantic Segmentation
• Towards Robust Semantic Correspondence: A Benchmark and Insights
• Vector Contrastive Learning For Pixel-Wise Pretraining In Medical Vision
• Do It Yourself: Learning Semantic Correspondence from Pseudo-Labels
• OV-COAST: Cost Aggregation with Optimal Transport for Open-Vocabulary Semantic Segmentation
• Semantic Correspondence: Unified Benchmarking and a Strong Baseline
• Statistical transformer networks: learning shape and appearance models via self supervision
• AerOSeg: Harnessing SAM for Open-Vocabulary Segmentation in Remote Sensing Images

Related

• Convolutional Hough Matching Networks
• Hierarchical Metric Learning and Matching for 2D and 3D Geometric Correspondences
• All Graphs Lead to Rome: Learning Geometric and Cycle-Consistent Representations with Graph Convolutional Networks
• Universal Correspondence Network
• Learning-based Natural Geometric Matching with Homography Prior