{"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/reliable-and-efficient-image-cropping-a-grid","title":"Reliable and Efficient Image Cropping: A Grid Anchor based Approach","arxiv_id":"1904.04441","date":"2019-04-09","proceeding":"CVPR 2019 6","authors":["Hui Zeng","Lida Li","Zisheng Cao","Lei Zhang"],"abstract":"Image cropping aims to improve the composition as well as aesthetic quality\nof an image by removing extraneous content from it. Existing image cropping\ndatabases provide only one or several human-annotated bounding boxes as the\ngroundtruth, which cannot reflect the non-uniqueness and flexibility of image\ncropping in practice. The employed evaluation metrics such as\nintersection-over-union cannot reliably reflect the real performance of\ncropping models, either. This work revisits the problem of image cropping, and\npresents a grid anchor based formulation by considering the special properties\nand requirements (e.g., local redundancy, content preservation, aspect ratio)\nof image cropping. Our formulation reduces the searching space of candidate\ncrops from millions to less than one hundred. Consequently, a grid anchor based\ncropping benchmark is constructed, where all crops of each image are annotated\nand more reliable evaluation metrics are defined. We also design an effective\nand lightweight network module, which simultaneously considers the region of\ninterest and region of discard for more accurate image cropping. Our model can\nstably output visually pleasing crops for images of different scenes and run at\na speed of 125 FPS. Code and dataset are available at:\nhttps://github.com/HuiZeng/Grid-Anchor-based-Image-Cropping.","url_abs":"http://arxiv.org/abs/1904.04441v1","url_pdf":"http://arxiv.org/pdf/1904.04441v1.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":"reliable-and-efficient-image-cropping-a-grid","repo_url":"https://github.com/HuiZeng/Grid-Anchor-based-Image-Cropping","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-cropping","task_name":"Image Cropping"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1904.04441","atlas_url":"https://app.syntology.ai/?focus=1904.04441","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}