# Improved Descriptors for Patch Matching and Reconstruction

## TL;DR

This paper introduces a convolutional neural network-based local image descriptor that significantly improves patch matching and 3D reconstruction, supported by a new comprehensive dataset and extensive evaluations.

## Contribution

It presents a multi-resolution ConvNet for learning local descriptors and a new large-scale dataset with diverse scenes and conditions, enhancing patch matching and reconstruction performance.

## Key findings

- Outperforms state-of-the-art descriptors in patch matching
- Achieves better 3D reconstruction accuracy
- Validated on multiple public datasets

## Abstract

We propose a convolutional neural network (ConvNet) based approach for learning local image descriptors which can be used for significantly improved patch matching and 3D reconstructions. A multi-resolution ConvNet is used for learning keypoint descriptors. We also propose a new dataset consisting of an order of magnitude more number of scenes, images, and positive and negative correspondences compared to the currently available Multi-View Stereo (MVS) [18] dataset. The new dataset also has better coverage of the overall viewpoint, scale, and lighting changes in comparison to the MVS dataset. We evaluate our approach on publicly available datasets, such as Oxford Affine Covariant Regions Dataset (ACRD) [12], MVS [18], Synthetic [6] and Strecha [15] datasets to quantify the image descriptor performance. Scenes from the Oxford ACRD, MVS and Synthetic datasets are used for evaluating the patch matching performance of the learnt descriptors while the Strecha dataset is used to evaluate the 3D reconstruction task. Experiments show that the proposed descriptor outperforms the current state-of-the-art descriptors in both the evaluation tasks.