Lightweight Real-Time Object Detection is available on Edge2, VIM3 and VIM4.
YOLOv7-Tiny is a Lightweight YOLO model. Compared with YOLOv8n, YOLOv7-Tiny has similar inference speed but theoretically higher accuracy due to more parameters. The fastest recognition speed can reach 30 frames per second.
We have released the Real-Time YOLOv7-Tiny Demo to help you get started more easily. The only thing you need to do is replace the model with your trained model. We provide two type of demos: Python and C++. You can choose one based on your preference or familiarity.
Find more details in our documentation.
Doc
Edge2 - YOLOv7-tiny Edge2 C++ Demo[Khadas Docs]
VIM3 - YOLOv7-tiny VIM3 C++ Demo[Khadas Docs]
VIM3 - YOLOv7-tiny KSNN Demo[Khadas Docs]
VIM4 - YOLOv7-tiny VIM4 C++ Demo[Khadas Docs]