Saved in:
Bibliographic Details
Main Authors: Thilakarathne, Haritha, Nibali, Aiden, He, Zhen, Morgan, Stuart
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2401.03262
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909063451770880
author Thilakarathne, Haritha
Nibali, Aiden
He, Zhen
Morgan, Stuart
author_facet Thilakarathne, Haritha
Nibali, Aiden
He, Zhen
Morgan, Stuart
contents Group activity recognition in video is a complex task due to the need for a model to recognise the actions of all individuals in the video and their complex interactions. Recent studies propose that optimal performance is achieved by individually tracking each person and subsequently inputting the sequence of poses or cropped images/optical flow into a model. This helps the model to recognise what actions each person is performing before they are merged to arrive at the group action class. However, all previous models are highly reliant on high quality tracking and have only been evaluated using ground truth tracking information. In practice it is almost impossible to achieve highly reliable tracking information for all individuals in a group activity video. We introduce an innovative deep learning-based group activity recognition approach called Rendered Pose based Group Activity Recognition System (RePGARS) which is designed to be tolerant of unreliable tracking and pose information. Experimental results confirm that RePGARS outperforms all existing group activity recognition algorithms tested which do not use ground truth detection and tracking information.
format Preprint
id arxiv_https___arxiv_org_abs_2401_03262
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Group Activity Recognition using Unreliable Tracked Pose
Thilakarathne, Haritha
Nibali, Aiden
He, Zhen
Morgan, Stuart
Computer Vision and Pattern Recognition
I.4.9
Group activity recognition in video is a complex task due to the need for a model to recognise the actions of all individuals in the video and their complex interactions. Recent studies propose that optimal performance is achieved by individually tracking each person and subsequently inputting the sequence of poses or cropped images/optical flow into a model. This helps the model to recognise what actions each person is performing before they are merged to arrive at the group action class. However, all previous models are highly reliant on high quality tracking and have only been evaluated using ground truth tracking information. In practice it is almost impossible to achieve highly reliable tracking information for all individuals in a group activity video. We introduce an innovative deep learning-based group activity recognition approach called Rendered Pose based Group Activity Recognition System (RePGARS) which is designed to be tolerant of unreliable tracking and pose information. Experimental results confirm that RePGARS outperforms all existing group activity recognition algorithms tested which do not use ground truth detection and tracking information.
title Group Activity Recognition using Unreliable Tracked Pose
topic Computer Vision and Pattern Recognition
I.4.9
url https://arxiv.org/abs/2401.03262