From General Actions to Domain-Specific Monitoring: Prior-Adaptive Transfer for Skeleton-Based Action Recognition
2026-07-03 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors look at how action recognition models that work well for general movements may struggle when used for specific tasks like monitoring health. They created a system called PATS that helps these models focus on the most important movements for a given task by removing less useful information. This makes the models work better and faster for things like spotting signs of Alzheimer's or detecting falls. Their method uses a special step that prunes and refines the model to keep only what is needed for the task.
Skeleton-based action recognitionTransfer learningMotion priorsModel compressionPruningSpatio-temporal backboneAlzheimer's detectionFall detectionAdaptive prior transferDomain adaptation
Authors
Hao Wang, Di Yang, Jiangtao Wang
Abstract
Skeleton-based action recognition models have recently shown strong performance on large-scale benchmarks with general actions. However, directly transferring them to domain-specific tasks e.g., healthcare monitoring, is often suboptimal, as such tasks are narrow in scope and may be relevant to only a subset of general motion priors. Moreover, not all pretrained motion patterns are equally useful for a specific task, and retaining less relevant components may hinder adaptation and increase computational cost. To address these challenges, we propose Prior-Adaptive Transfer of Skeletons (PATS), a framework that adapts general skeleton-based models by selectively retaining task-relevant motion priors while filtering redundant ones during transfer. PATS follows a standard pipeline that extracts skeleton signals from videos and employs a spatio-temporal backbone pre-trained on general actions. The key contribution lies in a novel Adaptive Prior Transfer module, which performs model compression as a prior selection mechanism through iterative pruning and refinement. Experiments on two specific action recognition tasks, Alzheimer's detection and fall detection, show consistent improvements in both performance and efficiency over competitive baselines. The code will be released upon acceptance.