CKD-LQPOSE: Towards a Real-World Low-quality Cross-Task Distilled Pose Estimation Architecture

Simple item page
dc.contributor.authorLiu, Taoen_US
dc.contributor.authorYao, Beijien_US
dc.contributor.authorHuang, Junen_US
dc.contributor.authorWang, Yaen_US
dc.contributor.editorChen, Renjieen_US
dc.contributor.editorRitschel, Tobiasen_US
dc.contributor.editorWhiting, Emilyen_US
dc.date.accessioned2024-10-13T18:05:44Z
dc.date.available2024-10-13T18:05:44Z
dc.date.issued2024
dc.description.abstractAlthough human pose estimation (HPE) methods have achieved promising results, they remain challenging in real-world lowquality (LQ) scenarios. Moreover, due to the general lack of modeling of LQ information in currently public HPE datasets, it is difficult to accurately evaluate the performance of the HPE methods in LQ scenarios. Hence, we propose a novel CKD-LQPose architecture, which is the first architecture fusing cross-task feature information in HPE that uses a cross-task distillation method to merge HPE information and well-quality (WQ) information. The CKD-LQPose architecture effectively enables adaptive feature learning from LQ images and improves their quality to enhance HPE performance. Additionally, we introduce the PatchWQ-Gan module to obtain WQ information and the refined transformer decoder (RTD) module to refine the features further. In the inference stage, CKD-LQPose removes the PatchWQ-Gan and RTD modules to reduce the computational burden. Furthermore, to accurately evaluate the HPE methods in LQ scenarios, we develop an RLQPose-DS test benchmark. Extensive experiments on RLQPose-DS, real-world images, and LQ versions of well-known datasets such as COCO, MPII, and CrowdPose demonstrate CKD-LQPose outperforms state-of-the-art approaches by a large margin, demonstrating its effectiveness in realworld LQ scenarios.en_US
dc.description.sectionheadersHuman II
dc.description.seriesinformationPacific Graphics Conference Papers and Posters
dc.identifier.doi10.2312/pg.20241321
dc.identifier.isbn978-3-03868-250-9
dc.identifier.pages12 pages
dc.identifier.urihttps://doi.org/10.2312/pg.20241321
dc.identifier.urihttps://diglib.eg.org/handle/10.2312/pg20241321
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: Computing methodologies → Image processing; Interest point and salient region detections
dc.subjectComputing methodologies → Image processing
dc.subjectInterest point and salient region detections
dc.titleCKD-LQPOSE: Towards a Real-World Low-quality Cross-Task Distilled Pose Estimation Architectureen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
pg20241321.pdf
Size:
7.82 MB
Format:
Adobe Portable Document Format
Download
Collections
PG2024 Conference Papers and Posters