{"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-to-predict-indoor-illumination-from","title":"Learning to Predict Indoor Illumination from a Single Image","arxiv_id":"1704.00090","date":"2017-04-01","proceeding":null,"authors":["Marc-André Gardner","Kalyan Sunkavalli","Ersin Yumer","Xiaohui Shen","Emiliano Gambaretto","Christian Gagné","Jean-François Lalonde"],"abstract":"We propose an automatic method to infer high dynamic range illumination from\na single, limited field-of-view, low dynamic range photograph of an indoor\nscene. In contrast to previous work that relies on specialized image capture,\nuser input, and/or simple scene models, we train an end-to-end deep neural\nnetwork that directly regresses a limited field-of-view photo to HDR\nillumination, without strong assumptions on scene geometry, material\nproperties, or lighting. We show that this can be accomplished in a three step\nprocess: 1) we train a robust lighting classifier to automatically annotate the\nlocation of light sources in a large dataset of LDR environment maps, 2) we use\nthese annotations to train a deep neural network that predicts the location of\nlights in a scene from a single limited field-of-view photo, and 3) we\nfine-tune this network using a small dataset of HDR environment maps to predict\nlight intensities. This allows us to automatically recover high-quality HDR\nillumination estimates that significantly outperform previous state-of-the-art\nmethods. Consequently, using our illumination estimates for applications like\n3D object insertion, we can achieve results that are photo-realistic, which is\nvalidated via a perceptual user study.","url_abs":"http://arxiv.org/abs/1704.00090v3","url_pdf":"http://arxiv.org/pdf/1704.00090v3.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":"lighting-estimation","task_name":"Lighting Estimation"}],"methods":[],"datasets_introduced":[{"slug":"laval-indoor-hdr-dataset","name":"Laval Indoor HDR Dataset","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.00090","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}