{"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/fpgadnn-co-design-an-efficient-design","title":"FPGA/DNN Co-Design: An Efficient Design Methodology for IoT Intelligence on the Edge","arxiv_id":"1904.04421","date":"2019-04-09","proceeding":null,"authors":["Cong Hao","Xiaofan Zhang","Yuhong Li","Sitao Huang","JinJun Xiong","Kyle Rupnow","Wen-mei Hwu","Deming Chen"],"abstract":"While embedded FPGAs are attractive platforms for DNN acceleration on\nedge-devices due to their low latency and high energy efficiency, the scarcity\nof resources of edge-scale FPGA devices also makes it challenging for DNN\ndeployment. In this paper, we propose a simultaneous FPGA/DNN co-design\nmethodology with both bottom-up and top-down approaches: a bottom-up\nhardware-oriented DNN model search for high accuracy, and a top-down FPGA\naccelerator design considering DNN-specific characteristics. We also build an\nautomatic co-design flow, including an Auto-DNN engine to perform\nhardware-oriented DNN model search, as well as an Auto-HLS engine to generate\nsynthesizable C code of the FPGA accelerator for explored DNNs. We demonstrate\nour co-design approach on an object detection task using PYNQ-Z1 FPGA. Results\nshow that our proposed DNN model and accelerator outperform the\nstate-of-the-art FPGA designs in all aspects including Intersection-over-Union\n(IoU) (6.2% higher), frames per second (FPS) (2.48X higher), power consumption\n(40% lower), and energy efficiency (2.5X higher). Compared to GPU-based\nsolutions, our designs deliver similar accuracy but consume far less energy.","url_abs":"http://arxiv.org/abs/1904.04421v1","url_pdf":"http://arxiv.org/pdf/1904.04421v1.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":"fpgadnn-co-design-an-efficient-design","repo_url":"https://github.com/TomG008/SkyNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"fpgadnn-co-design-an-efficient-design","repo_url":"https://github.com/jiangwx/SkyNet-ZCU104","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"fpgadnn-co-design-an-efficient-design","repo_url":"https://github.com/maxpark/SkyNet-1","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"c-code","task_name":"C++ code"},{"task_slug":null,"task_name":"GPU"},{"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/1904.04421","atlas_url":"https://app.syntology.ai/?focus=1904.04421","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}