PoseTrack Challenge

About the Workshop

Human poses and motions are important cues for analysis of videos with people and there is a strong evidence that representations based on body pose are highly effective for a variety of tasks such as activity recognition, content retrieval and social signal processing. In this workshop, we aim to further advance the state of the art by establishing a new large-scale benchmark for video-based human pose estimation and articulated tracking, and bringing together the community of researchers working on visual human analysis. The workshop will be organized around a challenge with three competition tracks focusing on single frame multi-person pose estimation, multi-person pose estimation in videos, and multi-person articulated tracking. To facilitate such benchmark and challenge we collect, annotate and release a new dataset that features videos with multiple people labeled with person tracks, articulated pose and occlusion labels. A centralized evaluation server will be set up to allow participants to evaluate on a held-out test set. We envision that the proposed benchmark will stimulate productive research both by providing a large and representative training dataset as well as providing a platform to objectively evaluate and compare the proposed methods.

What is Our Goal?

The main goals of this workshop will be to introduce the dataset and the related challenges, and to discuss the current state of the art as well as future directions to concentrate on. The planned panel discussion with world’s leading experts on this problem will be a fruitful input and source of ideas for all participants. As this will be the very first edition in the PoseTrack series, we as organizers also expect to receive valuable feedback from users and from the community on how to improve the benchmark. A few potential issues for the discussion are

  • Where are we in terms of solving the problem of human pose tracking?
  • What are positive and negative points of the PoseTrack dataset?
  • Are current evaluation metrics representative of the solution quality?
  • What are the major failure modes of current methods and how can these be eliminated?

Dataset Release
June 15

Challenge Opens
July 1

Challenge Deadline
July 28

Notification of Acceptance
August 15

Camera-Ready Submission
August 22

Workshop

October 23

Speakers

Michael Black

Michael Black

MPI for Intelligent Systems

Yaser Sheikh

Yaser Sheikh

Carnegie Mellon University

Jia Deng

Jia Deng

University of Michigan

Kevin Murphy

Kevin Murphy

Google

George Papandreou

George Papandreou

Google

Submit Your Contribution Work

The workshop will be organized around a challenge with three competition tracks focusing on

  • single frame multi-person pose estimation,
  • multi-person pose estimation in video, and
  • multi-person articulated tracking.

To participate in the PoseTrack 2017 Challenge you must:
Submit your results to the evaluation server, and
Submit a short or full-length paper describing your method.

Event Schedule

08:50 - 09:00IntroductionOrganizers
09:00 - 09:00Invited TalkJia Deng
09:30 - 10:00Invited TalkYaser Sheikh
10:00 - 10:25Challenge ResultsOrganizers
10:25 - 10:40Oral: Challenge WinnerTBD
10:40 - 11:00Coffee BreakEveryone
11:00 - 11:15Oral: Challenge Runner-upTBD
11:15 - 11:45Invited TalkK. Murphy & G. Papandreou
11:45 - 12:15Invited TalkMichael Black
12:15 - 12:30Closing Remarks & DiscussionOrganizers & Everyone

The schedule is subject to change

People

Organizers


Sponsored by

Oculus