Convolutional neural network architecture for geometric matching
This addresses geometric matching for computer vision applications, offering a novel method that improves generalization and performance without manual annotation.
The paper tackles the problem of determining correspondences between two images under geometric transformations like affine or thin-plate spline, proposing a convolutional neural network architecture that is trainable end-to-end and achieves state-of-the-art results on the Proposal Flow dataset.
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.