{"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/convolutional-image-captioning","title":"Convolutional Image Captioning","arxiv_id":"1711.09151","date":"2017-11-24","proceeding":"CVPR 2018 6","authors":["Jyoti Aneja","Aditya Deshpande","Alexander Schwing"],"abstract":"Image captioning is an important but challenging task, applicable to virtual\nassistants, editing tools, image indexing, and support of the disabled. Its\nchallenges are due to the variability and ambiguity of possible image\ndescriptions. In recent years significant progress has been made in image\ncaptioning, using Recurrent Neural Networks powered by long-short-term-memory\n(LSTM) units. Despite mitigating the vanishing gradient problem, and despite\ntheir compelling ability to memorize dependencies, LSTM units are complex and\ninherently sequential across time. To address this issue, recent work has shown\nbenefits of convolutional networks for machine translation and conditional\nimage generation. Inspired by their success, in this paper, we develop a\nconvolutional image captioning technique. We demonstrate its efficacy on the\nchallenging MSCOCO dataset and demonstrate performance on par with the\nbaseline, while having a faster training time per number of parameters. We also\nperform a detailed analysis, providing compelling reasons in favor of\nconvolutional language generation approaches.","url_abs":"http://arxiv.org/abs/1711.09151v1","url_pdf":"http://arxiv.org/pdf/1711.09151v1.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":"convolutional-image-captioning","repo_url":"https://github.com/aditya12agd5/convcap","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"convolutional-image-captioning","repo_url":"https://github.com/NaskyD/convnet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"convolutional-image-captioning","repo_url":"https://github.com/davinhill/Convolution_Captioning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"convolutional-image-captioning","repo_url":"https://github.com/xuewyang/Fashion_Captioning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"image-captioning","task_name":"Image Captioning"},{"task_slug":"text-generation","task_name":"Text Generation"},{"task_slug":"translation","task_name":"Translation"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.09151","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}