ICCV 2019 Workshop

Face Recognition in the Wild

Nov 2nd, 2019

Seoul, Korea

Call for papers

      Though almost saturated performance has been achieved on several classic face recognition tasks in academia, including LFW and Megaface, there are still many open problems for face recognition in industrial applications. For example, the training data might be quite noisy and imbalanced. Our workshop is mainly to discuss how to solve these problems. We call for papers in the following pecific subtopics and also welcome papers for general face recognition. The submission link is here .
  • Large-scale face recognition
  • Face recognition with imbalanced training data in the low-shot learning scenario
  • Generative model for face synthesis
  • Face representation learning
  • Disentangled representation learning
  • Training with noise labels
  • Image understanding with knowledge base
  • 3D analysis & synthesis
  • Expression analysis
  • How humans and face verification algorithms can work together
  • Bias in face recognition

Location

  • E3

News

  • 07/01/2019 We are going to host workshop "Face Recognition in the Wild" at ICCV 2019

Challenge of Recognizing People in the Real World

8:30-9:15
Delving into High Performance Detector for Finding Tiny Faces
Tang Xu
9:15-9:35
MT-DR Net: Multi-task Face Information Analysis
Yue Ming
9:35-10:00 Break;
10:00-10:20
A Practical Design for Face Recognition with Anti-Spoofing Based on Non-Visible Light Cameras
Songnan Xi, Lingbo Yang, Yao Zhao
10:30-11:15
Invited Talk: Uncertainty and Bias in Face Recognition and Expression Analysis
Weihong Deng
11:15-12:30
Poster Session
2:15-3:00
Invited Talk:"How to Regulate Face Recognition", by Prof. Erik Learned-Mille
Prof. Erik Learned-Miller
3:30-3:50 "Dyn-arcFace: Dynamic Additive Angular Margin Loss for Deep Face Recognition", Jichao Jiao, Jian Jiao, Weilun Liu
3:50-4:10 "Adjacent Agglomeration Face Detector", Yang Bai, Zhicheng Zhao, Fei Su, Hui Tian
4:10-5:00 Panel Discussion