{"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/compact-convolutional-neural-network-cascade","title":"Compact Convolutional Neural Network Cascade for Face Detection","arxiv_id":"1508.01292","date":"2015-08-06","proceeding":null,"authors":["Ilya Kalinovskii","Vladimir Spitsyn"],"abstract":"The problem of faces detection in images or video streams is a classical\nproblem of computer vision. The multiple solutions of this problem have been\nproposed, but the question of their optimality is still open. Many algorithms\nachieve a high quality face detection, but at the cost of high computational\ncomplexity. This restricts their application in the real-time systems. This\npaper presents a new solution of the frontal face detection problem based on\ncompact convolutional neural networks cascade. The test results on FDDB dataset\nshow that it is competitive with state-of-the-art algorithms. This proposed\ndetector is implemented using three technologies: SSE/AVX/AVX2 instruction sets\nfor Intel CPUs, Nvidia CUDA, OpenCL. The detection speed of our approach\nconsiderably exceeds all the existing CPU-based and GPU-based algorithms.\nBecause of high computational efficiency, our detector can processing 4K Ultra\nHD video stream in real time (up to 27 fps) on mobile platforms (Intel Ivy\nBridge CPUs and Nvidia Kepler GPUs) in searching objects with the dimension\n60x60 pixels or higher. At the same time its performance weakly dependent on\nthe background and number of objects in scene. This is achieved by the\nasynchronous computation of stages in the cascade.","url_abs":"http://arxiv.org/abs/1508.01292v3","url_pdf":"http://arxiv.org/pdf/1508.01292v3.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":"compact-convolutional-neural-network-cascade","repo_url":"https://github.com/Bkmz21/CompactCNNCascade","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"4k","task_name":"4k"},{"task_slug":null,"task_name":"CPU"},{"task_slug":"computational-efficiency","task_name":"Computational Efficiency"},{"task_slug":"face-detection","task_name":"Face Detection"},{"task_slug":null,"task_name":"GPU"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}