Can an Image Classifier Suffice For Action Recognition?

Quanfu Fan, Chun-Fu Chen, Rameswar Panda

Keywords: action recognition, image classifier, super image, vision transformer

Abstract: We explore a new perspective on video understanding by casting the video recognition problem as an image recognition task. Our approach rearranges input video frames into super images, which allow for training an image classifier directly to fulfill the task of action recognition, in exactly the same way as image classification. With such a simple idea, we show that transformer-based image classifiers alone can suffice for action recognition. In particular, our approach demonstrates strong and promising performance against SOTA methods on several public datasets including Kinetics400, Moments In Time, Something-Something V2 (SSV2), Jester and Diving48. We also experiment with the prevalent ResNet image classifiers in computer vision to further validate our idea. The results on both Kinetics400 and SSV2 are comparable to some of the best-performed CNN approaches based on spatio-temporal modeling. Our source codes and models are available at \url{https://github.com/IBM/sifar-pytorch}.

One-sentence Summary: We propose the idea of super images to re-purpose an image classifer for action recognition.