Hi Folks,
I have a small zoo of computer vision models implemented using the new neural network functionality in Mathematica v11:
AlexNet: similar to ImageIdentify, maybe of interest to those interested in understanding the internals of how a neural network does classification.
FaceDetection: Implements a face detector, similar to FindFaces, of interest to those who are interested in the internals.
Yolo: Implements Joseph Redmon's tiny YOLO model for object detection and localisation. Tiny YOLO recognises twenty object categories {"aeroplane","bicycle","bird","boat","bottle","bus","car","cat","chair","cow","diningtable","dog","horse","motorbike","person","pottedplant","sheep","sofa","train","tvmonitor"}:
Project is in a github repository: https://github.com/jfrancis71/CognitoZoo
The codebase and weights are all under an MIT licence which is fairly liberal, with the exception of the tiny YOLO weights file, which comes from Joseph Redmon (and that has a very different but also liberal license).
Please feel free to download and experiment with, and I hope it is of interest to the Mathematica community,
Kind regards,
Julian Francis.