Advanced topics in computer vision

Objectives

With a focus on the image classification task, the goal of this course is to go beyond large-scale and balanced data sets by considering data and computational resource constraints.

In the case of data constraints, this course will cover few-shot and one-shot image classification, long-tailed distribution, continual learning, and incremental learning. In the case of computation resource constraints, this course will cover model compression and distillation, weight quantization, binary neural networks, the training of quantified networks, and hashing methods for representation learning.

More information about this course is available HERE.