Continuity rotation representation for head pose estimation without keypoints

In this paper, we present an improved end-to-end head pose estimation method in an unconstrained environment, which transforms the Head Pose Estimation(HPE) problem into a problem of directly predicting continuous 6D rotation matrix parameters belongs 3D Special Orthogonal Group(SO(3)). The method u...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhao, Xiong, Sulaiman, Sarina, Chen, Liang, Dong, Min, Duo, Yunfeng, Song, Hao
Format: Conference or Workshop Item
Published: 2023
Subjects:
Online Access:http://eprints.utm.my/107865/
http://dx.doi.org/10.1145/3594315.3594341
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.utm.107865
record_format eprints
spelling my.utm.1078652024-10-08T06:41:47Z http://eprints.utm.my/107865/ Continuity rotation representation for head pose estimation without keypoints Zhao, Xiong Sulaiman, Sarina Chen, Liang Dong, Min Duo, Yunfeng Song, Hao QA75 Electronic computers. Computer science In this paper, we present an improved end-to-end head pose estimation method in an unconstrained environment, which transforms the Head Pose Estimation(HPE) problem into a problem of directly predicting continuous 6D rotation matrix parameters belongs 3D Special Orthogonal Group(SO(3)). The method uses RepVGGplusL2pse as the backbone, followed by one FC layer to output the results, model be trained on 300W-LP. The improved Root Mean Square Error of Geodesic Distance(RSME_GD) is used as the loss function to enhance the accuracy. The experiments on the two public face datasets AFLW-2000 and BIWI show that the results measured by Mean Absolute Error of Vectors (MAEV) are improved by 19.68% and 13.98% respectively compared with the original SOTA method. 2023 Conference or Workshop Item PeerReviewed Zhao, Xiong and Sulaiman, Sarina and Chen, Liang and Dong, Min and Duo, Yunfeng and Song, Hao (2023) Continuity rotation representation for head pose estimation without keypoints. In: 9th International Conference on Computing and Artificial Intelligence, ICCAI 2023, 17 March 2023-20 March 2023, Tianjin, China. http://dx.doi.org/10.1145/3594315.3594341
institution Universiti Teknologi Malaysia
building UTM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Teknologi Malaysia
content_source UTM Institutional Repository
url_provider http://eprints.utm.my/
topic QA75 Electronic computers. Computer science
spellingShingle QA75 Electronic computers. Computer science
Zhao, Xiong
Sulaiman, Sarina
Chen, Liang
Dong, Min
Duo, Yunfeng
Song, Hao
Continuity rotation representation for head pose estimation without keypoints
description In this paper, we present an improved end-to-end head pose estimation method in an unconstrained environment, which transforms the Head Pose Estimation(HPE) problem into a problem of directly predicting continuous 6D rotation matrix parameters belongs 3D Special Orthogonal Group(SO(3)). The method uses RepVGGplusL2pse as the backbone, followed by one FC layer to output the results, model be trained on 300W-LP. The improved Root Mean Square Error of Geodesic Distance(RSME_GD) is used as the loss function to enhance the accuracy. The experiments on the two public face datasets AFLW-2000 and BIWI show that the results measured by Mean Absolute Error of Vectors (MAEV) are improved by 19.68% and 13.98% respectively compared with the original SOTA method.
format Conference or Workshop Item
author Zhao, Xiong
Sulaiman, Sarina
Chen, Liang
Dong, Min
Duo, Yunfeng
Song, Hao
author_facet Zhao, Xiong
Sulaiman, Sarina
Chen, Liang
Dong, Min
Duo, Yunfeng
Song, Hao
author_sort Zhao, Xiong
title Continuity rotation representation for head pose estimation without keypoints
title_short Continuity rotation representation for head pose estimation without keypoints
title_full Continuity rotation representation for head pose estimation without keypoints
title_fullStr Continuity rotation representation for head pose estimation without keypoints
title_full_unstemmed Continuity rotation representation for head pose estimation without keypoints
title_sort continuity rotation representation for head pose estimation without keypoints
publishDate 2023
url http://eprints.utm.my/107865/
http://dx.doi.org/10.1145/3594315.3594341
_version_ 1814043543563403264
score 13.209306