Revisiting 3D ResNets for Video Recognition

By Xianzhi Du, Yeqing Li, Yin Cui, Jing Li, Irwan Bello. 2021

22 citations recorded by the source corpus.

A recent work from Bello shows that training and scaling strategies may be more significant than model architectures for visual recognition. This short note studies effective training and scaling strategies for video recognition models. We propose a simple scaling strategy for 3D ResNets, in combination with improved training strategies and minor architectural changes. The resulting models, termed 3D ResNet-RS, attain competitive performance of 81.0 on Kinetics-400 and 83.8 on Kinetics-600 without pre-training. When pre-trained on a large Web Video Text dataset, our best model achieves 83.5 and 84.3 on Kinetics-400 and Kinetics-600. The proposed scaling rule is further evaluated in a self-supervised setup using contrastive learning, demonstrating improved performance. Code is available at: https://github.com/tensorflow/models/tree/master/official.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Chen Sun
2. Anurag Arnab
3. Cordelia Schmid
4. Xuehan Xiong
5. Mi Zhang
6. Zhichao Lu
7. Shen Yan
8. Andrew Zisserman

In Computer Vision and Pattern Recognition

1. Cordelia Schmid
2. Anurag Arnab
3. Chen Sun
4. Luc Van Gool
5. Yi Yang
6. Xiaogang Wang
7. Andrew Zisserman
8. Jitendra Malik

Citations in the corpus

• Multiview Transformers for Video Recognition (2022, methodology): “We also improve upon R3D-RS [19], which also used WTS pre-training, by 6.4% and 6.0% on Kinetics-400 and -600.”