{"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/wide-slice-residual-networks-for-food","title":"Wide-Slice Residual Networks for Food Recognition","arxiv_id":"1612.06543","date":"2016-12-20","proceeding":null,"authors":["Niki Martinel","Gian Luca Foresti","Christian Micheloni"],"abstract":"Food diary applications represent a tantalizing market. Such applications,\nbased on image food recognition, opened to new challenges for computer vision\nand pattern recognition algorithms. Recent works in the field are focusing\neither on hand-crafted representations or on learning these by exploiting deep\nneural networks. Despite the success of such a last family of works, these\ngenerally exploit off-the shelf deep architectures to classify food dishes.\nThus, the architectures are not cast to the specific problem. We believe that\nbetter results can be obtained if the deep architecture is defined with respect\nto an analysis of the food composition. Following such an intuition, this work\nintroduces a new deep scheme that is designed to handle the food structure.\nSpecifically, inspired by the recent success of residual deep network, we\nexploit such a learning scheme and introduce a slice convolution block to\ncapture the vertical food layers. Outputs of the deep residual blocks are\ncombined with the sliced convolution to produce the classification score for\nspecific food categories. To evaluate our proposed architecture we have\nconducted experimental results on three benchmark datasets. Results demonstrate\nthat our solution shows better performance with respect to existing approaches\n(e.g., a top-1 accuracy of 90.27% on the Food-101 challenging dataset).","url_abs":"http://arxiv.org/abs/1612.06543v1","url_pdf":"http://arxiv.org/pdf/1612.06543v1.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":"wide-slice-residual-networks-for-food","repo_url":"https://github.com/fishba11/food-1010keras","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"wide-slice-residual-networks-for-food","repo_url":"https://github.com/lubagloukhova/food-101-keras","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"wide-slice-residual-networks-for-food","repo_url":"https://github.com/salimshaik/food-","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"wide-slice-residual-networks-for-food","repo_url":"https://github.com/stratospark/food-101-keras","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1612.06543","atlas_url":"https://app.syntology.ai/?focus=1612.06543","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}