{"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/salient-object-subitizing","title":"Salient Object Subitizing","arxiv_id":"1607.07525","date":"2016-07-26","proceeding":"CVPR 2015 6","authors":["Jianming Zhang","Shugao Ma","Mehrnoosh Sameki","Stan Sclaroff","Margrit Betke","Zhe Lin","Xiaohui Shen","Brian Price","Radomir Mech"],"abstract":"We study the problem of Salient Object Subitizing, i.e. predicting the\nexistence and the number of salient objects in an image using holistic cues.\nThis task is inspired by the ability of people to quickly and accurately\nidentify the number of items within the subitizing range (1-4). To this end, we\npresent a salient object subitizing image dataset of about 14K everyday images\nwhich are annotated using an online crowdsourcing marketplace. We show that\nusing an end-to-end trained Convolutional Neural Network (CNN) model, we\nachieve prediction accuracy comparable to human performance in identifying\nimages with zero or one salient object. For images with multiple salient\nobjects, our model also provides significantly better than chance performance\nwithout requiring any localization process. Moreover, we propose a method to\nimprove the training of the CNN subitizing model by leveraging synthetic\nimages. In experiments, we demonstrate the accuracy and generalizability of our\nCNN subitizing model and its applications in salient object detection and image\nretrieval.","url_abs":"http://arxiv.org/abs/1607.07525v1","url_pdf":"http://arxiv.org/pdf/1607.07525v1.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-retrieval","task_name":"Image Retrieval"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"salient-object-detection","task_name":"RGB Salient Object Detection"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"salient-object-detection-1","task_name":"Salient Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[{"slug":"salient-object-subitizing-dataset","name":"Salient Object Subitizing Dataset","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1607.07525","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}