Learning Two-View Correspondences and Geometry Using Order-Aware Network

Abstract

Establishing correspondences between two images requires both local and global spatial context. Given putative correspondences of feature points in two views, in this paper, we propose Order-Aware Network, which infers the probabilities of correspondences being inliers and regresses the relative pose encoded by the essential matrix. Specifically, this proposed network is built hierarchically and comprises three novel operations. First, to capture the local context of sparse correspondences, the network clusters unordered input correspondences by learning a soft assignment matrix. These clusters are in a canonical order and invariant to input permutations. Next, the clusters are spatially correlated to form the global context of correspondences. After that, the context-encoded clusters are recovered back to the original size through a proposed upsampling operator. We intensively experiment on both outdoor and indoor datasets. The accuracy of the two-view geometry and correspondences are significantly improved over the state-of-the-arts. Code will be available at https://github.com/zjhthu/OANet.git.

Accepted to ICCV 2019, and Winner solution to both tracks of CVPR IMW 2019 Challenge. Code will be available soon at https://github.com/zjhthu/OANet.git

References in corpus (8)

• Semi-Supervised Classification with Graph Convolutional Networks
• Inductive Representation Learning on Large Graphs
• Instance Normalization: The Missing Ingredient for Fast Stylization
• YFCC100M: The New Data in Multimedia Research
• DeMoN: Depth and Motion Network for Learning Monocular Stereo
• LF-Net: Learning Local Features from Images
• GeoDesc: Learning Local Descriptors by Integrating Geometry Constraints
• Self-Improving Visual Odometry

Cited by in corpus (10)

• DISK: Learning local features with policy gradient
• Reference Pose Generation for Long-term Visual Localization via Learned Features and View Synthesis
• LoFTR: Detector-Free Local Feature Matching with Transformers
• TopicFM: Robust and Interpretable Topic-Assisted Feature Matching
• Learning Second-Order Attentive Context for Efficient Correspondence Pruning
• Efficient Neighbourhood Consensus Networks via Submanifold Sparse Convolutions
• RANSAC-Flow: generic two-stage image alignment
• An End to End Network Architecture for Fundamental Matrix Estimation
• Descriptor-Free Multi-View Region Matching for Instance-Wise 3D Reconstruction
• Cascade Network with Guided Loss and Hybrid Attention for Two-view Geometry