{"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/a-4d-light-field-dataset-and-cnn","title":"A 4D Light-Field Dataset and CNN Architectures for Material Recognition","arxiv_id":"1608.06985","date":"2016-08-24","proceeding":null,"authors":["Ting-Chun Wang","Jun-Yan Zhu","Ebi Hiroaki","Manmohan Chandraker","Alexei A. Efros","Ravi Ramamoorthi"],"abstract":"We introduce a new light-field dataset of materials, and take advantage of\nthe recent success of deep learning to perform material recognition on the 4D\nlight-field. Our dataset contains 12 material categories, each with 100 images\ntaken with a Lytro Illum, from which we extract about 30,000 patches in total.\nTo the best of our knowledge, this is the first mid-size dataset for\nlight-field images. Our main goal is to investigate whether the additional\ninformation in a light-field (such as multiple sub-aperture views and\nview-dependent reflectance effects) can aid material recognition. Since\nrecognition networks have not been trained on 4D images before, we propose and\ncompare several novel CNN architectures to train on light-field images. In our\nexperiments, the best performing CNN architecture achieves a 7% boost compared\nwith 2D image classification (70% to 77%). These results constitute important\nbaselines that can spur further research in the use of CNNs for light-field\napplications. Upon publication, our dataset also enables other novel\napplications of light-fields, including object detection, image segmentation\nand view interpolation.","url_abs":"http://arxiv.org/abs/1608.06985v1","url_pdf":"http://arxiv.org/pdf/1608.06985v1.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-classification","task_name":"Image Classification"},{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"material-recognition","task_name":"Material Recognition"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"image-classification","task_name":"image-classification"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[{"slug":"light-field-material","name":"Light-Field Material","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1608.06985","atlas_url":"https://app.syntology.ai/?focus=1608.06985","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}