{"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/improving-object-counting-with-heatmap","title":"Improving Object Counting with Heatmap Regulation","arxiv_id":"1803.05494","date":"2018-03-14","proceeding":null,"authors":["Shubhra Aich","Ian Stavness"],"abstract":"In this paper, we propose a simple and effective way to improve one-look\nregression models for object counting from images. We use class activation map\nvisualizations to illustrate the drawbacks of learning a pure one-look\nregression model for a counting task. Based on these insights, we enhance\none-look regression counting models by regulating activation maps from the\nfinal convolution layer of the network with coarse ground-truth activation maps\ngenerated from simple dot annotations. We call this strategy heatmap regulation\n(HR). We show that this simple enhancement effectively suppresses false\ndetections generated by the corresponding one-look baseline model and also\nimproves the performance in terms of false negatives. Evaluations are performed\non four different counting datasets --- two for car counting (CARPK, PUCPR+),\none for crowd counting (WorldExpo) and another for biological cell counting\n(VGG-Cells). Adding HR to a simple VGG front-end improves performance on all\nthese benchmarks compared to a simple one-look baseline model and results in\nstate-of-the-art performance for car counting.","url_abs":"http://arxiv.org/abs/1803.05494v2","url_pdf":"http://arxiv.org/pdf/1803.05494v2.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":"improving-object-counting-with-heatmap","repo_url":"https://github.com/littleaich/heatmap-regulation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"improving-object-counting-with-heatmap","repo_url":"https://github.com/SajithMR/Counting_plant_parts","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"crowd-counting","task_name":"Crowd Counting"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-counting","task_name":"Object Counting"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"heatmap","method_name":"Heatmap"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}