AgentPose: Progressive Distribution Alignment via Feature Agent for Human Pose Distillation
Pose distillation is widely adopted to reduce model size in human pose estimation. However, existing methods primarily emphasize the transfer of teacher knowledge while often neglecting the performance degradation resulted from the curse of capacity gap between teacher and student. To address this issue, we propose AgentPose, a novel pose distillation method that integrates a feature agent to model the distribution of teacher features and progressively aligns the distribution of student features with that of the teacher feature, effectively overcoming the capacity gap and enhancing the ability of knowledge transfer. Our comprehensive experiments conducted on the COCO dataset substantiate the effectiveness of our method in knowledge transfer, particularly in scenarios with a high capacity gap.
Code (0)
등록된 구현이 없습니다.
Tasks
Pose EstimationTransfer LearningSimilar Papers 제목 키워드 기반
Semantic-Space-Intervened Diffusive Alignment for Visual Classification
Cross-modal alignment is an effective approach to improving visual classification. Existing studies typically enforce a one-step mapping that uses deep neural networks to project the visual features to mimic the distribu…
Classificationcross-modal alignmentView Distribution Alignment with Progressive Adversarial Learning for UAV Visual Geo-Localization
Unmanned Aerial Vehicle (UAV) visual geo-localization aims to match images of the same geographic target captured from different views, i.e., the UAV view and the satellite view. It is very challenging due to the large a…
geo-localizationBridging Modalities via Progressive Re-alignment for Multimodal Test-Time Adaptation
Test-time adaptation (TTA) enables online model adaptation using only unlabeled test data, aiming to bridge the gap between source and target distributions. However, in multimodal scenarios, varying degrees of distributi…
Test-time AdaptationContrastive LearningProgressive Multimodal Alignment for Continual Instruction Tuning
Multimodal Large Language Models (MLLMs) rely on a projector to align visual representations with the language embedding space, making it central to cross-modal understanding. In Multimodal Continual Instruction Tuning (…
Progressive Feature Alignment for Unsupervised Domain Adaptation
Unsupervised domain adaptation (UDA) transfers knowledge from a label-rich source domain to a fully-unlabeled target domain. To tackle this task, recent approaches resort to discriminative domain transfer in virtue of ps…
Domain AdaptationUnsupervised Domain Adaptation