From Keypoints to Object Landmarks via Self-Training Correspondence: A novel approach to Unsupervised Landmark Discovery

作     者：Mallis, Dimitrios Sanchez, Enrique Bell, Matt Tzimiropoulos, Georgios 

作者机构：The Computer Vision Lab University of Nottingham NG8 1BB United Kingdom Samsung AI Center Cambridge CB1 2RE United Kingdom The Department of Animal and Agriculture Hartpury University GL19 3BE United Kingdom The School of Electronic Engineering and Computer Science Queen Mary University of London E1 4NS United Kingdom 

出 版 物：《arXiv》 (arXiv)

年 卷 期：2022年

核心收录：

主　　题：Iterative methods 

摘      要：This paper proposes a novel paradigm for the unsupervised learning of object landmark detectors. Contrary to existing methods that build on auxiliary tasks such as image generation or equivariance, we propose a self-training approach where, departing from generic keypoints, a landmark detector and descriptor is trained to improve itself, tuning the keypoints into distinctive landmarks. To this end, we propose an iterative algorithm that alternates between producing new pseudo-labels through feature clustering and learning distinctive features for each pseudo-class through contrastive learning. With a shared backbone for the landmark detector and descriptor, the keypoint locations progressively converge to stable landmarks, filtering those less stable. Compared to previous works, our approach can learn points that are more flexible in terms of capturing large viewpoint changes. We validate our method on a variety of difficult datasets, including LS3D, BBCPose, Human3.6M and PennAction, achieving new state of the art results. Code and models can be found at https://***/dimitrismallis/KeypointsToLandmarks. Copyright © 2022, The Authors. All rights reserved.