Process This: Efficient object detection using Yolov5 and TDA4x processors | Video | TI.com (original) (raw)
Process This: Efficient object detection using Yolov5 and TDA4x processors
00:38:22 | 14 DEC 2021
In this Process This episode, learn how to optimize deep learning models for the deep learning hardware accelerator on TDA4x processors. We will use a Yolov5 example and show you how simple it is to modify the model architecture so that it runs efficiently and performs best on TDA4x processor without compromising accuracy.
Webinar topics:
- TI Model Zoo introduction
- Yolov5 model overview
- Optimizations for the deep learning hardware accelerator on TDA4x processors
- Model compilation using open-source run-time software
- TI Edge AI Cloud tool for free model bench-marking and evaluation