Machine Learning for Images Version 12 image processing and computer vision use extensively updated machine learning and neural net capabilities and introduce several built-in, high-level functions for object recognition, face analysis, restyling and more. In addition, with a growing number of pre-trained networks available from the Wolfram Neural Net Repository, one can use available pre-trained networks immediately, or manipulate and reassemble them to train on new data. Powerful network surgery capabilities enable transfer learning, which allows for solving problems using much smaller datasets. Built-in image style transfer using one or more templates. » Expanded built-in computer vision functions. » Automated object detection and recognition. » Significantly improved face detection. » Built-in facial age, gender, emotion estimation. » Immediately accessible trained and untrained neural nets. » Efficient and customizable image net encoders and decoders. » Extensive support for neural net dissection and reassembly. » Advanced neural net measurement and analysis. » GPU accelerated neural net training and evaluation. » Improved text recognition through customizable settings. » Related Examples Built-in Image Style Transfer » Style Transfer for Creative Art » Build-in Object Recognition » Object Recognition & Tracking in Videos » Improved Face Detection » Perform Face Interpolation » Simple Face Recognition » Facial 3D Reconstruction » A Model for Single-Image Depth Estimation » A Model to Estimate Geo Location from an Image » A Model for Super-Resolution » A Model to Translate Satellite Photos and Street Maps »
