Level: Master 1
Javier CIVERA, University of Zaragoza, Spain
Programming Language: Python
Textbooks or readings：
- Richard Szeliski, Computer Vision: Algorithms and Applications, Springer, 2010
- Richard Hartley and Andrew Zisserman, Multiple View Geometry in Computer Vision, Cambridge University Press, 2003.
- Ian Goodfellow, Yoshua Bengio and Aaron Courville, Deep Learning, MIT Press, 2016
The course will give an overview on some of the most relevant topics in computer vision, focusing in particular in 3D vision (the estimation of the camera motion and scene structure from images) and deep learning for visual tasks (object recognition or semantic segmentation among others). The theoretical details and formulation of the problems will be first detailed, and then the students will develop several practical assignments in order to acquire hands-on experience with the standard tools of the field.
What Will I Learn?
- Estimate the 6 degrees-of-freedom motion between several images of a scene
- Estimate the depth of a scene from several images of it
- Understand the geometric models of single view, two views and multiple views of a scene
- Understand the fundamentals of deep learning
- Know the main deep network architectures for computer vision problems
- Apply deep learning to vision-related problems (e.g., object recognition)
- Linera Algebra and Calculus at bachelor level