
Manifold-Aligned Graph Regularization (MAGR) addresses misalignment in feature replay for continual assessment challenges in Action Quality Assessment.


coremsg

magr-manifold-aligned-graph-regularization-for-continual-action-quality-assessment


MAGR: Manifold-Aligned Graph Regularization for Continual Action Quality Assessment


title_rewrite


The key to solving Continual Action Quality Assessment (Continual-AQA) is to sequentially learn a task-consistent score-discriminative feature distribution, where the latent features express a strong correlation with the score labels regardless of the task or action types.


continual-learning-for-action-quality-assessment-mitigating-forgetting-by-modeling-task-consistent-score-discriminative-feature-distributions


Continual Learning for Action Quality Assessment: Mitigating Forgetting by Modeling Task-Consistent Score-Discriminative Feature Distributions
