Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

Spatio-Temporal Adapter

A spatio-temporal adapter is a lightweight neural network module designed to adapt pre-trained static image models for video understanding tasks through parameter-efficient fine-tuning. By integrating spatial and temporal reasoning mechanisms into a compact architecture, the adapter can be inserted into the layers of a frozen visual backbone to capture dynamic motion and cross-frame relationships without modifying the core pre-trained weights. This modular approach allows models lacking native temporal awareness to process sequential video data while updating only a small fraction of parameters, significantly reducing computational and storage costs compared to full model retraining.

1 item

ST-Adapter: Parameter-Efficient Image-to-Video Transfer Learning

ST-Adapter: Parameter-Efficient Image-to-Video Transfer Learning

Junting Pan, Ziyi Lin, Xiatian Zhu, Jing Shao, Hongsheng Li

OrganizationsCentre for Perceptual and Interactive Intelligence (CPII)The Chinese University of Hong KongUniversity of Surrey

Why you should read this

Proposes a lightweight Spatio-Temporal Adapter that enables frozen pre-trained image vision transformers to perform video action recognition by updating only about eight percent of parameters while matching or exceeding full fine-tuning performance.

Capitalizing on large pre-trained models for various downstream tasks of interest have recently emerged with promising performance. Due to the ever-growing model size, the standard full fine-tuning based task adaptation strategy becomes prohibitively costly in terms of model training and storage. This has led to a new research direction in parameter-efficient transfer learning. However, existing attempts typically focus on downstream tasks from the same modality (e.g., image understanding) of the pre-trained model. This creates a limit because in some specific modalities, (e.g., video understanding) such a strong pre-trained model with sufficient knowledge is less or not available. In this work, we investigate such a novel cross-modality transfer learning setting, namely parameter-efficient image-to-video transfer learning. To solve this problem, we propose a new Spatio-Temporal Adapter (ST-Adapter) for parameter-efficient fine-tuning per video task. With a built-in spatio-temporal reasoning capability in a compact design, ST-Adapter enables a pre-trained image model without temporal knowledge to reason about dynamic video content at a small (~8%) per-task parameter cost, requiring approximately 20 times fewer updated parameters compared to previous work. Extensive experiments on video action recognition tasks show that our ST-Adapter can match or even outperform the strong full fine-tuning strategy and state-of-the-art video models, whilst enjoying the advantage of parameter efficiency. Code and model are available at https://github.com/linziyi96/st-adapter

Added

2026-09-26