{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/revisiting-3d-resnets-for-video-recognition","title":"Revisiting 3D ResNets for Video Recognition","arxiv_id":"2109.01696","date":"2021-09-03","proceeding":null,"authors":["Xianzhi Du","Yeqing Li","Yin Cui","Rui Qian","Jing Li","Irwan Bello"],"abstract":"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.","url_abs":"https://arxiv.org/abs/2109.01696v1","url_pdf":"https://arxiv.org/pdf/2109.01696v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"revisiting-3d-resnets-for-video-recognition","repo_url":"https://github.com/tensorflow/models","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null},{"paper_slug":"revisiting-3d-resnets-for-video-recognition","repo_url":"https://github.com/2023-MindSpore-1/ms-code-216/tree/main/resnet3d","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"revisiting-3d-resnets-for-video-recognition","repo_url":"https://github.com/MindSpore-paper-code-2/code2/tree/main/resnet3d","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"revisiting-3d-resnets-for-video-recognition","repo_url":"https://github.com/MindSpore-paper-code-3/code5/tree/main/resnet3d","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"revisiting-3d-resnets-for-video-recognition","repo_url":"https://github.com/code-implementation1/Code7/tree/main/resnet3d","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"action-classification","task_name":"Action Classification"},{"task_slug":"contrastive-learning","task_name":"Contrastive Learning"},{"task_slug":"video-recognition","task_name":"Video Recognition"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"3d-convolution","method_name":"3D Convolution"},{"method_slug":"3d-resnet-rs","method_name":"3D ResNet-RS"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"bottleneck-residual-block","method_name":"Bottleneck Residual Block"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"resnet-d","method_name":"ResNet-D"},{"method_slug":"resnet-rs","method_name":"ResNet-RS"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"squeeze-and-excitation-block","method_name":"Squeeze-and-Excitation Block"},{"method_slug":"stochastic-depth","method_name":"Stochastic Depth"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"xavier-initialization","method_name":"Xavier Initialization"}],"datasets_introduced":[],"methods_introduced":[{"slug":"3d-resnet-rs","name":"3D ResNet-RS","full_name":"3D ResNet-RS"}],"results":[{"leaderboard":"/sota/action-classification-on-kinetics-400","task":"Action Classification","dataset":"Kinetics-400","model":"R3D-RS-200","rank_in_archive_order":101,"of":207,"metrics":{"Acc@1":"80.4","Acc@5":"94.4"},"uses_additional_data":false},{"leaderboard":"/sota/action-classification-on-kinetics-600","task":"Action Classification","dataset":"Kinetics-600","model":"R3D-RS-200","rank_in_archive_order":41,"of":65,"metrics":{"Top-1 Accuracy":"83.1"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2109.01696","atlas_url":"https://app.syntology.ai/?focus=2109.01696","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}