{"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/tiny-dsod-lightweight-object-detection-for","title":"Tiny-DSOD: Lightweight Object Detection for Resource-Restricted Usages","arxiv_id":"1807.11013","date":"2018-07-29","proceeding":null,"authors":["Yuxi Li","Jiuwei Li","Weiyao Lin","Jianguo Li"],"abstract":"Object detection has made great progress in the past few years along with the\ndevelopment of deep learning. However, most current object detection methods\nare resource hungry, which hinders their wide deployment to many resource\nrestricted usages such as usages on always-on devices, battery-powered low-end\ndevices, etc. This paper considers the resource and accuracy trade-off for\nresource-restricted usages during designing the whole object detection\nframework. Based on the deeply supervised object detection (DSOD) framework, we\npropose Tiny-DSOD dedicating to resource-restricted usages. Tiny-DSOD\nintroduces two innovative and ultra-efficient architecture blocks: depthwise\ndense block (DDB) based backbone and depthwise feature-pyramid-network (D-FPN)\nbased front-end. We conduct extensive experiments on three famous benchmarks\n(PASCAL VOC 2007, KITTI, and COCO), and compare Tiny-DSOD to the\nstate-of-the-art ultra-efficient object detection solutions such as Tiny-YOLO,\nMobileNet-SSD (v1 & v2), SqueezeDet, Pelee, etc. Results show that Tiny-DSOD\noutperforms these solutions in all the three metrics (parameter-size, FLOPs,\naccuracy) in each comparison. For instance, Tiny-DSOD achieves 72.1% mAP with\nonly 0.95M parameters and 1.06B FLOPs, which is by far the state-of-the-arts\nresult with such a low resource requirement.","url_abs":"http://arxiv.org/abs/1807.11013v1","url_pdf":"http://arxiv.org/pdf/1807.11013v1.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":"tiny-dsod-lightweight-object-detection-for","repo_url":"https://github.com/lyxok1/Tiny-DSOD","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1807.11013","atlas_url":"https://app.syntology.ai/?focus=1807.11013","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}