PECoP: Parameter Efficient Continual Pretraining for Action Quality Assessment

Amirhossein Dadashzadeh, Shuchao Duan, Alan Whone, Majid Mirmehdi; Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024, pp. 42-52

The limited availability of labelled data in Action Quality Assessment (AQA), has forced previous works to fine-tune their models pretrained on large-scale domain-general datasets. This common approach results in weak generalisation, particularly when there is a significant domain shift. We propose a novel, parameter efficient, continual pretraining framework, PECoP, to reduce such domain shift via an additional pretraining stage. In PECoP, we introduce 3D-Adapters, inserted into the pretrained model, to learn spatiotemporal, in-domain information via self-supervised learning where only the adapter modules' parameters are updated. We demonstrate PECoP's ability to enhance the performance of recent state-of-the-art methods (MUSDL, CoRe, and TSA) applied to AQA, leading to considerable improvements on benchmark datasets, JIGSAWS (| 6.0%), MTL-AQA (| 0.99%), and FineDiving (| 2.54%). We also present a new Parkinson's Disease dataset, PD4T, of real patients performing four various actions, where we surpass (| 3.56%) the state-of-the-art in comparison. Our code, pretrained models, and the PD4T dataset are available at https://github.com/Plrbear/PECoP.