Learning with Limited and Imperfect Data

L2ID @ ECCV 2022 Introduction

Boyi Li: SITTA: Single Image Texture Translation for Data Augmentation

Boshen Zhang: Learning from Noisy Labels with Coarse-to-Fine Sample Credibility Modeling

Sukmin Yun: OpenCoS: Contrastive Semi-supervised Learning for Handling Open-set Unlabeled Data

Holger Caesar: Autonomous vehicles from imperfect and limited labels

Yu Cheng: Towards data efficient vision-language (VL) models

SangYun Lee: Learning Multiple Probabilistic Degradation Generators for Unsupervised Superresolution

Rabab Abdelfattah: PLMCL: Partial-Label Momentum Curriculum Learning for Multi-label Classification

Nir Zabari: Open-Vocabulary Semantic Segmentation using Test-Time Distillation

Niv Cohen: "This is my unicorn, Fluffy": Personalizing frozen vision-language representations

Hamza Khan: Timestamp-Supervised Action Segmentation with Graph Convolutional Networks

Jiageng Zhu: SW-VAE: Weakly Supervised Learn Disentangled Representation Via Latent Factor Swapping

Niv Cohen: Out-of-Distribution Detection Without Class Labels

Andong Tan: Unsupervised Adaptive Object Detection with Class Label Shift Weighted Local Features

Leonid Karlinsky: Different facets of limited supervision

Vadim Sushko: One-Shot Synthesis of Images and Segmentation Masks

Bharath Hariharan: When life gives you lemons: Making lemonade from limited labels

Abhay Rawat: Semi-Supervised Domain Adaptation by Similarity based Pseudo-label Injection

Ruiwen Li: TransCAM: Transformer Attention-based CAM Refinement for Weakly Supervised Segmentation

Ishan Misra: General purpose visual recognition across modalities with limited supervision

Sharon Li: How to Handle Data Shifts? Challenges, Research Progress and Path Forward

Xiuye Gu: Open-Vocabulary Detection and Segmentation

Yinfei Yang: Learning Visual and Vision-Language Model With Noisy Image Text Pairs

L2ID CVPR2021 Panel Sessions - Part 2

L2ID CVPR2021 Panel Sessions - Part 3

L2ID CVPR2021 Panel Sessions - Part 1

A Causal View of Compositional Zero Shot Recognition

Cluster-driven Graph Federated Learning over Multiple Domains

Rethinking Ensemble Distillation for Semantic Segmentation Based Unsupervised Domain Adaptation

ProFeat: Unsupervised Image Clustering via Progressive Feature Refinement