02:08

L2ID @ ECCV 2022 Introduction

19:59

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

07:00

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

13:32

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

27:18

Holger Caesar: Autonomous vehicles from imperfect and limited labels

30:31

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

06:45

SangYun Lee: Learning Multiple Probabilistic Degradation Generators for Unsupervised Superresolution

14:51

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

19:48

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

20:00

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

06:50

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

18:31

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

19:25

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

20:20

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

24:49

Leonid Karlinsky: Different facets of limited supervision

16:09

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

29:11

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

08:19

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

12:00

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

25:44

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

31:33

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

27:24

Xiuye Gu: Open-Vocabulary Detection and Segmentation

22:34

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

02:25:04

L2ID CVPR2021 Panel Sessions - Part 2

01:53:42

L2ID CVPR2021 Panel Sessions - Part 3

01:56:06

L2ID CVPR2021 Panel Sessions - Part 1

10:00

A Causal View of Compositional Zero Shot Recognition

06:09

Cluster-driven Graph Federated Learning over Multiple Domains

09:14

Rethinking Ensemble Distillation for Semantic Segmentation Based Unsupervised Domain Adaptation

10:01

ProFeat: Unsupervised Image Clustering via Progressive Feature Refinement

08:39

One Shot GAN learning to generate samples from Single Images and Videos

15:46

Unlocking the Full Potential of Small Datawith Diverse Supervision

12:38

A Simple Framework for Cross-Domain Few-Shot Recognition with Unlabeled Data

09:25

Weak Multi-View Supervision for Surface Mapping Estimation

10:28

ReMP Rectified Metric Propagation for Few Shot Learning

04:37

BalaGAN: Cross-Modal Image Translation Between Imbalanced Domains

25:15

Dina Katabi - Making Contrastive Learning Robust to Shortcuts and Generalize it to New Modalities

22:22

Sanja Fidler - Image GANs for Reducing Pixel-Wise Supervision

10:05

Efficient Pre trained Features and Recurrent Pseudo Labeling in Unsupervised Domain Adaptation

08:36

One shot action recognition in challenging therapy scenarios

20:56

Humphrey Shi - Escaping the Big Data Paradigm with Compact Transformers

17:36

Anurag Arnab - Video Understanding with Imperfect Data

18:37

Trevor Darrell - Recent Progress on Unsupervised Detection and Adaptation

13:34

Unsupervised Discriminative Embedding for Sub Action Learning in Complex Activities

17:09

Learning Unbiased Representations via Mutual Information Backpropagation

10:01

Domain Agnostic Score Based Meta Learning

07:13

Boosting Co teaching with Compression Regularization for Label Noise Poster

09:17

Adversarial Open Domain Adaption for Sketch to Photo Synthesis

09:01

Batch Normalization Embeddings for Deep Domain Generalization

09:23

Improving Semi Supervised Domain Adaptation Using Effective Target Selection and Semantics

10:01

Learning From Incomplete Features by Simultaneous Training of Neural Networks and Sparse Coding

09:36

TAEN Temporal aware embedding network for few shot action recognition

08:09

Training Rare Object Detection in Satellite Imagery with Synthetic GAN Images

20:52

Olga Russakovsky - Mitigating bias and privacy concerns in visual data

16:02

Training Deep Generative Models in Highly Incomplete Data Scenarios with Prior Regularization

15:10

Shot in the Dark Few Shot Learning with No Base Class Labels

18:59

PLM Partial Label Masking for Imbalanced Multi label Classification

14:47

Distill on the Go Online knowledge distillation in self supervised learning

15:56

Contrastive Learning Improves Model Robustness Under Label Noise

08:54

An Exploration into why Output Regularization Mitigates Label Noise