3D CNNs with Adaptive Temporal Feature Resolutions
CVPR 2021
Mohsen Fayyaz
Emad Bahrami
Ali Diba
Mehdi Noroozi
Ehsan Adeli
Luc Van Gool
Juergen Gall


While state-of-the-art 3D Convolutional Neural Networks (CNN) achieve very good results on action recognition datasets, they are computationally very expensive and require many GFLOPs. While the GFLOPs of a 3D CNN can be decreased by reducing the temporal feature resolution within the network, there is no setting that is optimal for all input clips. In this work, we therefore introduce a differentiable Similarity Guided Sampling (SGS) module, which can be plugged into any existing 3D CNN architecture. SGS empowers 3D CNNs by learning the similarity of temporal features and grouping similar features together. As a result, the temporal feature resolution is not anymore static but it varies for each input video clip. By integrating SGS as an additional layer within current 3D CNNs, we can convert them into much more efficient 3D CNNs with adaptive temporal feature resolutions (ATFR) . Our evaluations show that the proposed module improves the stateof-the-art by reducing the computational cost (GFLOPs) by half while preserving or even improving the accuracy. We evaluate our module by adding it to multiple state-ofthe-art 3D CNNs on various datasets such as Kinetics600, Kinetics-400, mini-Kinetics, Something-Something V2, UCF101, and HMDB51.



Paper and Supplementary Material

Fayyaz M.*, Bahrami E.*, Diba A., Noroozi M., Adeli E., Van Gool L., Gall J.
3D CNNs with Adaptive Temporal Feature Resolutions
CVPR, 2021.(hosted on ArXiv)

* indicates equal contribution



This template was originally made by Phillip Isola and Richard Zhang for a colorful ECCV project; the code can be found here.