{"id":56543,"date":"2024-07-02T06:00:00","date_gmt":"2024-07-02T13:00:00","guid":{"rendered":"https:\/\/learnopencv.com\/?p=56543"},"modified":"2025-02-26T22:22:08","modified_gmt":"2025-02-27T06:22:08","slug":"object-detection-on-edge-device","status":"publish","type":"post","link":"https:\/\/learnopencv.com\/object-detection-on-edge-device\/","title":{"rendered":"Object Detection on Edge Device: Deploying YOLOv8 on Luxonis OAK-D-Lite – Pothole Datset"},"content":{"rendered":"\n

Performing Object Detection on edge device<\/strong> is an exciting area for tech enthusiasts where we can implement powerful computer vision applications in compact, efficient packages. Here we show one such interesting embedded computer vision application by deploying models on a popular edge AI device like OAK-D-Lite<\/strong>.<\/p>\n\n\n

\"object-detection-edge-device-oakd\"<\/figure>\n\n\n

According to the World Health Organization’s <\/strong>2023 report, road accidents claim 1.19 million lives annually and cause non-fatal injuries to 20 to 50 million people. A major contributor to these unsafe driving conditions is road potholes. To address this issue, we demonstrate the procedure for running a fine-tuned pothole detection model (YOLOv8<\/strong>) on the OAK-D-Lite device. In a broader perspective, this deployment can serve as a baseline reference, which can be extended to assist drivers by providing real-time monitoring and warning indicators, thereby enhancing road safety.<\/p>\n\n\n\n

To learn more about OAK-D and its essential applications, we recommend that you bookmark our series on Stereo <\/a>Depth<\/a> and OAK-D<\/a>. <\/p>\n\n\n\n

By this article’s end, you can convert and deploy recent lightweight YOLO models on an OAK-D-Lite<\/strong> device.<\/p>\n\n\n\n

YOLO Master Post –  Every Model Explained<\/strong><\/h2>\n\n\n\n
Unlock the full story behind all the YOLO models’ evolutionary journey: Dive into our extensive pillar post, where we unravel the evolution from YOLOv1 to YOLOv10. This essential guide is packed with insights, comparisons, and a deeper understanding that you won’t find anywhere else.
Don’t miss out on this comprehensive resource, Mastering All Yolo Models<\/a> for a richer, more informed perspective on the YOLO series.<\/div>
<\/div><\/div>

Mastering All YOLO Models from YOLOv1 to YOLOv10: Papers Explained (2024)<\/a><\/h2><\/small><\/div><\/div>\n\n\n\n
    \n
  1. An Intro to OAK-D-Lite<\/a><\/li>\n\n\n\n
  2. DepthAI Pipeline for OAK-D-Lite<\/a><\/li>\n\n\n\n
  3. How Does OpenVINO IR Format Optimizes OAK-D-Lite?<\/a><\/li>\n\n\n\n
  4. Understanding the Pothole Dataset<\/a> <\/li>\n\n\n\n
  5. Code Walkthrough<\/a>\n
      \n
    1. Model Conversion: YOLO Pytorch to MyriadX Blob Format<\/a><\/li>\n\n\n\n
    2. OAK-D-Lite Deployment<\/a><\/li>\n<\/ol>\n<\/li>\n\n\n\n
    3. Video Inference Results on OAK-D-Lite<\/a><\/li>\n\n\n\n
    4. Key Takeaways of OAK-D Deployment<\/a><\/li>\n\n\n\n
    5. Conclusion<\/a><\/li>\n\n\n\n
    6. References<\/a><\/li>\n<\/ol>\n\n\n\n

      An Intro to OAK-D-Lite<\/h2>\n\n\n\n
      \n