PAPER 1
FrameExit: Conditional Early Exiting for Efficient Video Recognition
Amir Ghodrati
Babak Ehteshami Bejnordi
Amirhossein Habibian
(Qualcomm AI Research)
CVPR 2021 oral
Summary
The goal of the work is to achieve efficient action recognition in videos. The FrameExit framework adjusts the amount of computation based on the difficulty of the input, which can significantly reduce the computational requirements. More specifically, FrameExit is an efficient video recognition model that performs automatic early exiting by adjusting the computational budget on a per-video basis. This can significantly reduce the computational requirements when processing a video.
Citation
@inproceedings{ghodrati2021, title={FrameExit: Conditional Early Exiting for Efficient Video Recognition}, author={Ghodrati, Amir and Bejnordi, Babak Ehteshami and Habibian, Amirhossein}, booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition}, year={2021}}
Results
The researchers found that a few frames are sufficient for classifying most simple videos, while more difficult sample videos need more frames for detailed information. Using FrameExit, the model’s performance for a holistic video classification task yields 2.5x less MACs while maintaining accuracy. For a video classification task, FrameExit uses 1.3x to 5x less GFLOPs while maintaining accuracy.
Looking for more papers with code?
*Qualcomm AI Research is an initiative of Qualcomm Technologies, Inc.
