Decoupled Prompt-Adapter Tuning for Continual Activity Recognition
This addresses the need for models that continuously adapt in domains like surveillance and healthcare, though it appears incremental as it builds on existing prompt and adapter techniques.
The paper tackled the problem of continual action recognition by proposing Decoupled Prompt-Adapter Tuning (DPAT), which integrates adapters and learnable prompts to adapt to new video data without forgetting previous knowledge, achieving state-of-the-art performance on benchmarks.
Action recognition technology plays a vital role in enhancing security through surveillance systems, enabling better patient monitoring in healthcare, providing in-depth performance analysis in sports, and facilitating seamless human-AI collaboration in domains such as manufacturing and assistive technologies. The dynamic nature of data in these areas underscores the need for models that can continuously adapt to new video data without losing previously acquired knowledge, highlighting the critical role of advanced continual action recognition. To address these challenges, we propose Decoupled Prompt-Adapter Tuning (DPAT), a novel framework that integrates adapters for capturing spatial-temporal information and learnable prompts for mitigating catastrophic forgetting through a decoupled training strategy. DPAT uniquely balances the generalization benefits of prompt tuning with the plasticity provided by adapters in pretrained vision models, effectively addressing the challenge of maintaining model performance amidst continuous data evolution without necessitating extensive finetuning. DPAT consistently achieves state-of-the-art performance across several challenging action recognition benchmarks, thus demonstrating the effectiveness of our model in the domain of continual action recognition.