{"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/inverting-visual-representations-with","title":"Inverting Visual Representations with Convolutional Networks","arxiv_id":"1506.02753","date":"2015-06-09","proceeding":"CVPR 2016 6","authors":["Alexey Dosovitskiy","Thomas Brox"],"abstract":"Feature representations, both hand-designed and learned ones, are often hard\nto analyze and interpret, even when they are extracted from visual data. We\npropose a new approach to study image representations by inverting them with an\nup-convolutional neural network. We apply the method to shallow representations\n(HOG, SIFT, LBP), as well as to deep networks. For shallow representations our\napproach provides significantly better reconstructions than existing methods,\nrevealing that there is surprisingly rich information contained in these\nfeatures. Inverting a deep network trained on ImageNet provides several\ninsights into the properties of the feature representation learned by the\nnetwork. Most strikingly, the colors and the rough contours of an image can be\nreconstructed from activations in higher network layers and even from the\npredicted class probabilities.","url_abs":"http://arxiv.org/abs/1506.02753v4","url_pdf":"http://arxiv.org/pdf/1506.02753v4.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":"inverting-visual-representations-with","repo_url":"https://github.com/JepsonWong/CNN_Visualization","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"inverting-visual-representations-with","repo_url":"https://github.com/nsom/conv_inv","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1506.02753","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}