Self Supervised Correspondence

Self-supervised correspondence learning aims to establish accurate mappings between data points (e.g., pixels in images, points in point clouds) without relying on manually labeled correspondences. Current research focuses on developing robust methods that handle noisy data, variations in viewpoint or object pose, and diverse data modalities (images, point clouds, speech). This involves leveraging techniques like optimal transport, geometric consistency constraints, and self-reconstruction losses within various model architectures, including transformers and deep neural networks. The resulting advancements have significant implications for numerous applications, including 3D reconstruction, visual odometry, object pose estimation, and cross-modal retrieval.

Papers

July 8, 2024

SGOR: Outlier Removal by Leveraging Semantic and Geometric Information for Robust Point Cloud Registration
Guiyu Zhao, Zhentao Guo, Hongbin Ma
Point Cloud Registration Semantic Description Geometric Information Robust Registration Self Supervised Correspondence

April 17, 2024

Learning SO(3)-Invariant Semantic Correspondence via Local Shape Transform
Chunghyun Park, Seungwook Kim, Jaesik Park, Minsu Cho
Semantic Correspondence Global Shape Shape Descriptor Self Supervised Correspondence 3D Correspondence

March 10, 2024

SCORE: Self-supervised Correspondence Fine-tuning for Improved Content Representations
Amit Meghanani, Thomas Hain
Automatic Speech Recognition Supervised Fine Tuning Semantic Representation Speech Model Correspondence Learning Self Supervised Correspondence

December 4, 2023

iMatching: Imperative Correspondence Learning
Zitong Zhan, Dasong Gao, Yun-Jou Lin, Youjie Xia, Chen Wang
Correspondence Learning Self Supervised Correspondence Feature Correspondence

October 26, 2023

Cross-modal Active Complementary Learning with Self-refining Correspondence
Yang Qin, Yuan Sun, Dezhong Peng, Joey Tianyi Zhou, Xi Peng, Peng Hu
Cross Modal Self Supervised Correspondence Noisy Correspondence Image Text Benchmark

September 26, 2023

Doduo: Learning Dense Visual Correspondence from Unsupervised Semantic-Aware Flow
Zhenyu Jiang, Hanwen Jiang, Yuke Zhu
Unsupervised Setting Dense Correspondence Dense Visual Correspondence Self Supervised Correspondence

December 6, 2022

Self-Supervised Correspondence Estimation via Multiview Registration
Mohamed El Banani, Ignacio Rocco, David Novotny, Andrea Vedaldi, Natalia Neverova, Justin Johnson, Benjamin Graham
Correspondence Learning Consecutive Frame RGB D Sequence Self Supervised Correspondence Correspondence Estimation Multiview Registration

November 25, 2022

SCOOP: Self-Supervised Correspondence and Optimization-Based Scene Flow
Itai Lang, Dror Aiger, Forrester Cole, Shai Avidan, Michael Rubinstein
Ground Truth Flow Based Scene Flow Scene Flow Estimation Joint Entity Self Supervised Correspondence Scooping Strategy

October 13, 2022

Self-Supervised Geometric Correspondence for Category-Level 6D Object Pose Estimation in the Wild
Kaifeng Zhang, Yang Fu, Shubhankar Borse, Hong Cai, Fatih Porikli, Xiaolong Wang
Computer Vision Wild Challenge 6D Object Category Level 6D Object Pose Self Supervised Correspondence Canonical Shape

March 27, 2022

Locality-Aware Inter-and Intra-Video Reconstruction for Self-Supervised Correspondence Learning
Liulei Li, Tianfei Zhou, Wenguan Wang, Lu Yang, Jianwu Li, Yi Yang
Pixel Level Locality Sensitive Video Reconstruction Self Supervised Video Self Supervised Correspondence

March 2, 2022

Weakly Supervised Correspondence Learning
Zihan Wang, Zhangjie Cao, Yilun Hao, Dorsa Sadigh
Weak Supervision Correspondence Learning Self Supervised Correspondence Action Pair

December 29, 2021

COTReg:Coupled Optimal Transport based Point Cloud Registration
Guofeng Mei, Xiaoshui Huang, Litao Yu, Jian Zhang, Mohammed Bennamoun
Point Cloud Optimal Transport Point Cloud Registration Self Supervised Correspondence Matching Framework

December 9, 2021

GAN-Supervised Dense Visual Alignment
William Peebles, Jun-Yan Zhu, Richard Zhang, Antonio Torralba, Alexei A. Efros, Eli Shechtman
GAN Model GAN Training Self Supervised Correspondence