This chapter contains recipes for:
- Representing images as tensors/blobs
- Loading Deep Learning models from Caffe, Torch, and TensorFlow formats
- Getting input and output tensors' shapes for all layers
- Preprocessing images and inference in convolutional networks
- Measuring inference time and contributions to it from each layer
- Classifying images with GoogleNet/Inception and ResNet models
- Detecting objects with the Single Shot Detection (SSD) model
- Segmenting a scene using the Fully Convolutional Network (FCN) model
- Face detection using Single Shot Detection (SSD) and the ResNet model
- Prediction age and gender
