Self-supervised learning for multi-person pose estimation and tracking in videos

Self-supervised learning in video involves learning representations without using high-cost labels, which allows us to utilise the vast amount of free unlabelled videos available. As videos are particularly rich in spatio-temporal information, the representations learnt are useful in many downstrea...

Full description

Bibliographic Details
Main Author: Tan, Jia Min
Other Authors: Li Boyang
Format: Student Research Poster
Language:English
Published: Nanyang Technological University 2023
Subjects:
Online Access:https://hdl.handle.net/10356/170731