# DeepPoint3D: Learning Discriminative Local Descriptors using Deep Metric Learning on 3D Point Clouds

## TL;DR

This paper introduces DeepPoint3D, a novel deep learning approach that directly processes unstructured 3D point clouds to learn discriminative local descriptors without intermediate representations, outperforming existing methods.

## Contribution

It proposes a permutation-invariant deep network with a multi-margin contrastive loss for learning 3D local descriptors directly from point clouds, avoiding reliance on RGB-D or multi-view data.

## Key findings

- Outperforms state-of-the-art methods in 3D point matching
- Demonstrates effectiveness across various applications
- Achieves superior accuracy in local descriptor learning

## Abstract

Learning local descriptors is an important problem in computer vision. While there are many techniques for learning local patch descriptors for 2D images, recently efforts have been made for learning local descriptors for 3D points. The recent progress towards solving this problem in 3D leverages the strong feature representation capability of image based convolutional neural networks by utilizing RGB-D or multi-view representations. However, in this paper, we propose to learn 3D local descriptors by directly processing unstructured 3D point clouds without needing any intermediate representation. The method constitutes a deep network for learning permutation invariant representation of 3D points. To learn the local descriptors, we use a multi-margin contrastive loss which discriminates between similar and dissimilar points on a surface while also leveraging the extent of dissimilarity among the negative samples at the time of training. With comprehensive evaluation against strong baselines, we show that the proposed method outperforms state-of-the-art methods for matching points in 3D point clouds. Further, we demonstrate the effectiveness of the proposed method on various applications achieving state-of-the-art results.