In this course, you will :
- Investigate image classification, segmentation, object localization, and detection. Apply transfer learning to object detection and localization.
- Use object detection models such as regional-CNN and ResNet-50 to detect, localise, and label your own rubber duck images, as well as customise existing models and build your own models.
- Use image segmentation variants of the fully convolutional network (FCN), such as U-Net, and d) Mask-RCNN to identify and detect numbers, pets, zombies, and other objects.
- Using class activation maps and saliency maps, identify which parts of an image are being used by your model to make predictions, and then apply these ML interpretation methods to inspect and improve the design of a well-known network, AlexNet.
- Introduction to Computer Vision
- Object Detection
- Image Segmentation
- Visualization and Interpretability