{"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/self-supervised-spatiotemporal-feature","title":"Self-Supervised Spatiotemporal Feature Learning via Video Rotation Prediction","arxiv_id":"1811.11387","date":"2018-11-28","proceeding":null,"authors":["Longlong Jing","Xiaodong Yang","Jingen Liu","YingLi Tian"],"abstract":"The success of deep neural networks generally requires a vast amount of\ntraining data to be labeled, which is expensive and unfeasible in scale,\nespecially for video collections. To alleviate this problem, in this paper, we\npropose 3DRotNet: a fully self-supervised approach to learn spatiotemporal\nfeatures from unlabeled videos. A set of rotations are applied to all videos,\nand a pretext task is defined as prediction of these rotations. When\naccomplishing this task, 3DRotNet is actually trained to understand the\nsemantic concepts and motions in videos. In other words, it learns a\nspatiotemporal video representation, which can be transferred to improve video\nunderstanding tasks in small datasets. Our extensive experiments successfully\ndemonstrate the effectiveness of the proposed framework on action recognition,\nleading to significant improvements over the state-of-the-art self-supervised\nmethods. With the self-supervised pre-trained 3DRotNet from large datasets, the\nrecognition accuracy is boosted up by 20.4% on UCF101 and 16.7% on HMDB51\nrespectively, compared to the models trained from scratch.","url_abs":"http://arxiv.org/abs/1811.11387v2","url_pdf":"http://arxiv.org/pdf/1811.11387v2.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":[],"tasks":[{"task_slug":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":"prediction","task_name":"Prediction"},{"task_slug":"self-supervised-action-recognition","task_name":"Self-Supervised Action Recognition"},{"task_slug":"action-recognition","task_name":"Temporal Action Localization"},{"task_slug":"video-understanding","task_name":"Video Understanding"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"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":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/self-supervised-action-recognition-on-hmdb51","task":"Self-Supervised Action Recognition","dataset":"HMDB51","model":"3D RotNet (3D ResNet-18)","rank_in_archive_order":42,"of":48,"metrics":{"Frozen":"false","Pre-Training Dataset":"Kinetics400","Top-1 Accuracy":"33.7"},"uses_additional_data":false},{"leaderboard":"/sota/self-supervised-action-recognition-on-ucf101","task":"Self-Supervised Action Recognition","dataset":"UCF101","model":"3D RotNet (3D ResNet-18)","rank_in_archive_order":45,"of":53,"metrics":{"3-fold Accuracy":"62.9","Frozen":"false","Pre-Training Dataset":"Kinetics400"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.11387","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}