{"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/learning-audio-video-modalities-from-image","title":"Learning Audio-Video Modalities from Image Captions","arxiv_id":"2204.00679","date":"2022-04-01","proceeding":null,"authors":["Arsha Nagrani","Paul Hongsuck Seo","Bryan Seybold","Anja Hauth","Santiago Manen","Chen Sun","Cordelia Schmid"],"abstract":"A major challenge in text-video and text-audio retrieval is the lack of large-scale training data. This is unlike image-captioning, where datasets are in the order of millions of samples. To close this gap we propose a new video mining pipeline which involves transferring captions from image captioning datasets to video clips with no additional manual effort. Using this pipeline, we create a new large-scale, weakly labelled audio-video captioning dataset consisting of millions of paired clips and captions. We show that training a multimodal transformed based model on this data achieves competitive performance on video retrieval and video captioning, matching or even outperforming HowTo100M pretraining with 20x fewer clips. We also show that our mined clips are suitable for text-audio pretraining, and achieve state of the art results for the task of audio retrieval.","url_abs":"https://arxiv.org/abs/2204.00679v1","url_pdf":"https://arxiv.org/pdf/2204.00679v1.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":"image-captioning","task_name":"Image Captioning"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"video-captioning","task_name":"Video Captioning"},{"task_slug":"video-retrieval","task_name":"Video Retrieval"},{"task_slug":"zero-shot-video-retrieval","task_name":"Zero-Shot Video Retrieval"},{"task_slug":"zero-shot-text-to-audio-retrieval","task_name":"Zero-shot Text to Audio Retrieval"}],"methods":[],"datasets_introduced":[{"slug":"videocc3m","name":"VideoCC3M","full_name":"Video-Conceptual-Captions"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/zero-shot-video-retrieval-on-msr-vtt","task":"Zero-Shot Video Retrieval","dataset":"MSR-VTT","model":"A. Nagrani et. al.","rank_in_archive_order":33,"of":41,"metrics":{"text-to-video R@1":"19.4","text-to-video R@10":"50.3","text-to-video R@5":"39.5"},"uses_additional_data":true}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2204.00679","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}