{"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/deepsentibank-visual-sentiment-concept","title":"DeepSentiBank: Visual Sentiment Concept Classification with Deep Convolutional Neural Networks","arxiv_id":"1410.8586","date":"2014-10-30","proceeding":null,"authors":["Tao Chen","Damian Borth","Trevor Darrell","Shih-Fu Chang"],"abstract":"This paper introduces a visual sentiment concept classification method based\non deep convolutional neural networks (CNNs). The visual sentiment concepts are\nadjective noun pairs (ANPs) automatically discovered from the tags of web\nphotos, and can be utilized as effective statistical cues for detecting\nemotions depicted in the images. Nearly one million Flickr images tagged with\nthese ANPs are downloaded to train the classifiers of the concepts. We adopt\nthe popular model of deep convolutional neural networks which recently shows\ngreat performance improvement on classifying large-scale web-based image\ndataset such as ImageNet. Our deep CNNs model is trained based on Caffe, a\nnewly developed deep learning framework. To deal with the biased training data\nwhich only contains images with strong sentiment and to prevent overfitting, we\ninitialize the model with the model weights trained from ImageNet. Performance\nevaluation shows the newly trained deep CNNs model SentiBank 2.0 (or called\nDeepSentiBank) is significantly improved in both annotation accuracy and\nretrieval performance, compared to its predecessors which mainly use binary SVM\nclassification models.","url_abs":"http://arxiv.org/abs/1410.8586v1","url_pdf":"http://arxiv.org/pdf/1410.8586v1.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":"deepsentibank-visual-sentiment-concept","repo_url":"https://github.com/ColumbiaDVMM/ColumbiaImageSearch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1410.8586","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}