Papers › AIM: Adapting Image Models for Efficient Video Action Recognition

AIM: Adapting Image Models for Efficient Video Action Recognition

6 Feb 2023arXiv:2302.03024archive 2025-07-28

Taojiannan Yang, Yi Zhu, Yusheng Xie, Aston Zhang, Chen Chen, Mu Li

Recent vision transformer based video models mostly follow the ``image pre-training then finetuning" paradigm and have achieved great success on multiple video benchmarks. However, full finetuning such a video model could be computationally expensive and unnecessary, given the pre-trained image transformer models have demonstrated exceptional transferability. In this work, we propose a novel method to Adapt pre-trained Image Models (AIM) for efficient video understanding. By freezing the pre-trained image model and adding a few lightweight Adapters, we introduce spatial adaptation, temporal adaptation and joint adaptation to gradually equip an image model with spatiotemporal reasoning capability. We show that our proposed AIM can achieve competitive or even better performance than prior arts with substantially fewer tunable parameters on four video action recognition benchmarks. Thanks to its simplicity, our method is also generally applicable to different image pre-trained models, which has the potential to leverage more powerful image foundation models in the future. The project webpage is \url{https://adapt-image-models.github.io/}.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Action ClassificationAction RecognitionTemporal Action LocalizationVideo Understanding

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Action Classification Kinetics-400 AIM (CLIP ViT-L/14, 32x224) Acc@1 87.5 #35 of 207 Archive leaderboard report
Action Classification Kinetics-400 AIM (CLIP ViT-L/14, 32x224) Acc@5 97.7 #35 of 207 Archive leaderboard report
Action Classification Kinetics-700 AIM (CLIP ViT-L/14, 32x224) Top-1 Accuracy 80.4 #14 of 36 Archive leaderboard report
Action Recognition Diving-48 AIM (CLIP ViT-L/14, 32x224) Accuracy 90.6 #3 of 18 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AttentionDense ConnectionsLayer NormalizationLinear LayerMulti-Head AttentionResidual ConnectionSoftmaxVision Transformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections