{"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/avt-audio-video-transformer-for-multimodal","title":"AVT: Audio-Video Transformer for Multimodal Action Recognition","arxiv_id":null,"date":"2022-09-22","proceeding":"Submitted to ICLR 2022 9","authors":["Wentao Zhu","Jingru Yi","Kevin Hsu","Xiaohang Sun","Xiang Hao","Linda Liu","Mohamed Omar"],"abstract":"Action recognition is an essential field for video understanding. To learn from heterogeneous data sources effectively, in this work, we propose a novel multimodal action recognition approach termed Audio-Video Transformer (AVT). AVT uses a combination of video and audio signals to improve action recognition accuracy, leveraging the effective spatio-temporal representation by the video Transformer. For multimodal fusion, simply concatenating multimodal tokens in a cross-modal Transformer requires large computational and memory resources, instead we reduce the cross-modality complexity through an audio-video bottleneck Transformer. To improve the learning efficiency of multimodal Transformer, we integrate self-supervised objectives, i.e., audio-video contrastive learning, audio-video matching, and masked audio and video learning, into AVT training, which maps diverse audio and video representations into a common multimodal representation space. We further propose a masked audio segment loss to learn semantic audio activities in AVT. Extensive experiments and ablation studies on three public datasets and two in-house datasets consistently demonstrate the effectiveness of the proposed AVT. Specifically, AVT outperforms its previous state-of-the-art counterparts on Kinetics-Sounds and Epic-Kitchens-100 datasets by 8% and 1%, respectively, without external training data. AVT also surpasses one of the previous state-of-the-art video Transformers by 10% on the VGGSound dataset by leveraging the audio signal. Compared to one of the previous state-of-the-art multimodal Transformers, AVT is 1.3x more efficient in terms of FLOPs and improves the accuracy by 4.2% on Epic-Kitchens-100. Visualization results further demonstrate that the audio provides complementary and discriminative features, and our AVT can effectively understand the action from a combination of audio and video.","url_abs":"https://openreview.net/pdf?id=yFuHxmSwGus","url_pdf":"https://openreview.net/pdf?id=yFuHxmSwGus","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":"audio-classification","task_name":"Audio Classification"},{"task_slug":"contrastive-learning","task_name":"Contrastive Learning"},{"task_slug":"multi-modal-classification","task_name":"Multi-modal Classification"},{"task_slug":"video-understanding","task_name":"Video Understanding"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/action-recognition-on-epic-kitchens-100","task":"Action Recognition","dataset":"EPIC-KITCHENS-100","model":"AVT","rank_in_archive_order":15,"of":32,"metrics":{"Action@1":"47.2","Noun@1":"59.3","Verb@1":"70.4"},"uses_additional_data":false},{"leaderboard":"/sota/audio-classification-on-vggsound","task":"Audio Classification","dataset":"VGGSound","model":"AVT (Audio-Visual)","rank_in_archive_order":11,"of":23,"metrics":{"Top 1 Accuracy":"63.9","Top 5 Accuracy":"85.0"},"uses_additional_data":false},{"leaderboard":"/sota/audio-classification-on-vggsound","task":"Audio Classification","dataset":"VGGSound","model":"AVT (V)","rank_in_archive_order":19,"of":23,"metrics":{"Top 1 Accuracy":"53.2","Top 5 Accuracy":"74.8"},"uses_additional_data":false},{"leaderboard":"/sota/multi-modal-classification-on-vgg-sound","task":"Multi-modal Classification","dataset":"VGG-Sound","model":"AVT","rank_in_archive_order":4,"of":4,"metrics":{"Top-1 Accuracy":"63.9","Top-5 Accuracy":"85.0"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}