The Machine Perception Group at the School of Computer Science and Engineering develops algorithms and methods that will allow computers to reach the remarkable performance of humans, and in some cases even surpass human ability.
The Computer Vision Group’s current focus includes the analysis of multiple images, which entails computing a scene’s 3D structure from multiple views; improving image quality; video summarization; image segmentation and classification; visual recognition; visual learning; indexing into image and video databases; and visual perception, or psychophysics.
The Group’s breakthrough research has led to the creation of new companies and commercial applications in a wide range of industries, including transportation (camera-based systems for driver assistance, cruise control, and lane-change indicators); measurement of automobile parts; and applications for the entertainment, graphics, design and security industries.
Introduction to digital image processing: description of the imaging process, and learning basic concepts and operations.
The course is meant to provide the student an introductory working knowledge of modern neural networks that are in use in computer vision and image processing.
An introduction to deep learning, the course will describe various common network architectures and tools for applying them in different fields of CS
The course includes a wide overview of advanced deep learning topics. Additionally, it also includes a research project.
Introduction to Computer Graphics. The course covers geometric transformations, color and shading of surfaces, curves, rendering, and more.
Computational photography is a research area where image processing and computer vision meet computer graphics to yield tools that extend the boundaries of conventional photography.
Understanding Human Vision from a computational perspective while performing computational analysis of visual illusions and implementing computer vision algorithms.
The course will present the fundamental computational models behind scene interpretation, motion understanding and object recognition.
Lectures will be given by invited guest speakers. In addition, students who participate in the seminar will present their research area or papers from the professional literature.
Selected state-of-the-art papers from computer graphics and computational photography will be presented and discussed.
Students will be exposed to state-of-the-art research in 3D representation and generation, typically papers published in the recent top conferences in the field.