The goal of this course is to acquaint the students with the field of computer vision from a deep learning perspective. The students will learn how to analyze, design, implement, and evaluate any complex computer vision problem. The course covers both image and video processing, including image classification, object detection, object tracking, action recognition, image stylization, and synthetic data generation.
The specific objectives of this course are:
• Understand various architectures of Convolutional Neural Networks for image classification, object detection, video analysis, and synthetic visual data generation.
•Solve and analyze a Computer Vision problem using a specific theoretical apparatus.
•Understand and develop efficient fine-tuning strategies for increasing the performance of Convolutional Neural Networks with applications in the Computer Vision field.
•Understand the metrics used to evaluate complex networks, as well as visualizing the features learned by the networks.
- Teacher: Diana Borza