{"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/how-convolutional-neural-network-see-the","title":"How convolutional neural network see the world - A survey of convolutional neural network visualization methods","arxiv_id":"1804.11191","date":"2018-04-30","proceeding":null,"authors":["Zhuwei Qin","Fuxun Yu","ChenChen Liu","Xiang Chen"],"abstract":"Nowadays, the Convolutional Neural Networks (CNNs) have achieved impressive\nperformance on many computer vision related tasks, such as object detection,\nimage recognition, image retrieval, etc. These achievements benefit from the\nCNNs outstanding capability to learn the input features with deep layers of\nneuron structures and iterative training process. However, these learned\nfeatures are hard to identify and interpret from a human vision perspective,\ncausing a lack of understanding of the CNNs internal working mechanism. To\nimprove the CNN interpretability, the CNN visualization is well utilized as a\nqualitative analysis method, which translates the internal features into\nvisually perceptible patterns. And many CNN visualization works have been\nproposed in the literature to interpret the CNN in perspectives of network\nstructure, operation, and semantic concept. In this paper, we expect to provide\na comprehensive survey of several representative CNN visualization methods,\nincluding Activation Maximization, Network Inversion, Deconvolutional Neural\nNetworks (DeconvNet), and Network Dissection based visualization. These methods\nare presented in terms of motivations, algorithms, and experiment results.\nBased on these visualization methods, we also discuss their practical\napplications to demonstrate the significance of the CNN interpretability in\nareas of network design, optimization, security enhancement, etc.","url_abs":"http://arxiv.org/abs/1804.11191v2","url_pdf":"http://arxiv.org/pdf/1804.11191v2.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":"how-convolutional-neural-network-see-the","repo_url":"https://github.com/justinbellucci/cnn-visualizations-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"interpretability","method_name":"Interpretability"},{"method_slug":"network-dissection","method_name":"Network Dissection"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}