/Tiny-Yolo-3

Google Colab Notebook for creating and testing a Tiny Yolo 3 real-time object detection model.

Primary LanguageJupyter Notebook

Tiny-Yolo-3

Google Colab Notebook for creating and testing a Tiny Yolo 3 real-time object detection model. This notebook manually creates the Tiny Yolo 3 model layer by layer allowing it to be customized for the constraints of your hardware. This approach exposes what is "under the hood" of the tiny-yolo architecture. This notebook can be run to create an .h5 file that can be downloaded from the notebook.

Companion YouTube Video Tutorial: https://youtu.be/SWrTS63zgoU

Like/Follow the Black Magic AI Facebook page for the latest post about this project and others.

Image of Before and After results using this Tiny Yolo 3 Model

Image of Before and After results using this Tiny Yolo 3 Model

Reference

@article{YOLOv3,  
  title={YOLOv3: An Incremental Improvement},  
  author={J Redmon, A Farhadi },
  year={2018}
url={https://pjreddie.com/media/files/papers/YOLOv3.pdf}

Redmon, Joseph. “Yolo Web: Real-Time Object Detection.” YOLO: Real-Time Object Detection, Joseph Redmon, 2018, pjreddie.com/darknet/yolo/.

Based on code from Xiaochus, Larry. “YOLOv3” Github code, Xiaochus, Larry, 2018, https://github.com/xiaochus/YOLOv3.