Deep Regression on Manifolds: A 3D Rotation Case Study
Author:
- 1 December 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 166-174
Abstract
Many machine learning problems involve regressing variables on a non-Euclidean manifold–e.g. a discrete probability distribution, or the 6D pose of an object. One way to tackle these problems through gradient-based learning is to use a differentiable function that maps arbitrary inputs of a Euclidean space onto the manifold. In this paper, we establish a set of desirable properties for such mapping, and in particular highlight the importance of pre-images connectivity/convexity. We illustrate these properties with a case study regarding 3D rotations. Through theoretical considerations and methodological experiments on a variety of tasks, we review various differentiable mappings on the 3D rotation space, and conjecture about the importance of their local linearity. We show that a mapping based on Procrustes orthonormalization generally performs best among the mappings considered, but that a rotation vector representation might also be suitable when restricted to small angles.Keywords
All Related Versions
This publication has 22 references indexed in Scilit:
- Defining the Pose of Any 3D Rigid Object and an Associated DistanceInternational Journal of Computer Vision, 2017
- Image-Based Localization Using LSTMs for Structured Feature Correlation2017
- Geometric Deep Learning: Going beyond Euclidean dataIEEE Signal Processing Magazine, 2017
- Unsupervised Learning of Depth and Ego-Motion from Video2017
- Deep Residual Learning for Image Recognition2016
- A Novel Representation of Parts for Accurate 3D Object Detection and Tracking in Monocular Images2015
- PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization2015
- Multimodal templates for real-time detection of texture-less objects in heavily cluttered scenes2011
- Least-squares estimation of transformation parameters between two point patternsIEEE Transactions on Pattern Analysis and Machine Intelligence, 1991
- A Generalized Solution of the Orthogonal Procrustes ProblemPsychometrika, 1966