AII2022 - Competition

Challenge: Human Pose Estimation. To participate visit Kaggle at

About The Problem and the Competition:

    Human pose classification is one of the challenging computer vision problems which has potential applications in various fields such as visual surveillance, behavioral analysis, assisted living, and intelligent driver assistance system [1, 2].

    To handle more variety in human poses, we are using a fine-grained hierarchical pose classification dataset named Yoga-82 for this competition proposed by [1, 2]. This dataset formulates the pose estimation as a classification task. The Yoga-82 database has samples belonging to 82 different classes (yoga poses). The dataset provides hierarchical labels for yoga poses based on the body configuration of the pose. The dataset contains a three-level hierarchy including body positions, variations in body positions, and the actual pose names. There are 6, 20, and 82 classes in the first, second, and third levels, respectively.

Important Dates:
  • Submission deadline : 1st May 2022 (No extension).
  • Winners Announcement: 2nd May 2022.
  • AII2022 in Reggio Calabria, Italy : 1st – 3rd September 2022

    The top three winners of the competition will receive:
  • Free registration to the conference.
  • 14 days extra to submit their paper.
  • Certificate with secured position in the ranking list (i.e., 1st, 2nd, and 3rd).


  1. Verma, Manisha, Sudhakar Kumawat, Yuta Nakashima, and Shanmuganathan Raman. "Yoga-82: a new dataset for fine-grained classification of human poses." In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, pp. 1038-1039. 2020.
  2. The Yoga82 database homepage,