CaFlow: Enhancing Long-Term Action Quality Assessment with Causal Counterfactual Flow
This work addresses the problem of robustly assessing action quality over extended periods for applications like sports analysis and skill evaluation, representing an incremental improvement with novel method integration.
The paper tackles the challenge of long-term Action Quality Assessment (AQA) in videos, such as for sports or rehabilitation, by proposing CaFlow, a framework that integrates causal counterfactual de-confounding with bidirectional temporal modeling, achieving state-of-the-art performance on multiple benchmarks.
Action Quality Assessment (AQA) predicts fine-grained execution scores from action videos and is widely applied in sports, rehabilitation, and skill evaluation. Long-term AQA, as in figure skating or rhythmic gymnastics, is especially challenging since it requires modeling extended temporal dynamics while remaining robust to contextual confounders. Existing approaches either depend on costly annotations or rely on unidirectional temporal modeling, making them vulnerable to spurious correlations and unstable long-term representations. To this end, we propose CaFlow, a unified framework that integrates counterfactual de-confounding with bidirectional time-conditioned flow. The Causal Counterfactual Regularization (CCR) module disentangles causal and confounding features in a self-supervised manner and enforces causal robustness through counterfactual interventions, while the BiT-Flow module models forward and backward dynamics with a cycle-consistency constraint to produce smoother and more coherent representations. Extensive experiments on multiple long-term AQA benchmarks demonstrate that CaFlow achieves state-of-the-art performance. Code is available at https://github.com/Harrison21/CaFlow