Humans perceive the three-dimensional structure of the world with apparent ease. However, despite all of the recent advances in computer vision research, the dream of having a computer interpret an image at the same level as a two-year old remains elusive. Why is computer vision such a challenging problem and what is the current state of the art? Computer Vision: Algorithms and Applications explores the variety of techniques commonly used to analyze and interpret images. It also describes challenging real-world applications where vision is being successfully used, both for specialized applications such as medical imaging, and for fun, consumer-level tasks such as image editing and stitching, which students can apply to their own personal photos and videos. More than just a source of recipes, this exceptionally authoritative and comprehensive textbook/reference also takes a scientific approach to basic vision problems, formulating physical models of the imaging process before inverting them to produce descriptions of a scene. These problems are also analyzed using statistical models and solved using rigorous engineering techniques Topics and features: structured to support active curricula and project-oriented courses, with tips in the Introduction for using the book in a variety of customized courses; presents exercises at the end of each chapter with a heavy emphasis on testing algorithms and containing numerous suggestions for small mid-term projects; provides additional material and more detailed mathematical topics in the Appendices, which cover linear algebra, numerical techniques, and Bayesian estimation theory; suggests additional reading at the end of each chapter, including the latest research in each sub-field, in addition to a full Bibliography at the end of the book; supplies supplementary course material for students at the associated website, http://szeliski.org/Book/. Suitable for an upper-level undergraduate or graduate-level course in computer science or engineering, this textbook focuses on basic techniques that work under real-world conditions and encourages students to push their creative boundaries. Its design and exposition also make it eminently suitable as a unique reference to the fundamental techniques and current research literature in computer vision.

/pdf/computer-vision-algorithms-and-applications-25dn6wu83j.pdf

Computer Vision: Algorithms and Applications

We introduce a new iterative regularization procedure for inverse problems based on the use of Bregman distances, with particular focus on problems arising in image processing. We are motivated by the problem of restoring noisy and blurry images via variational methods by using total variation regularization. We obtain rigorous convergence results and effective stopping criteria for the general procedure. The numerical results for denoising appear to give significant improvement over standard models, and preliminary results for deblurring/denoising are very encouraging.

/pdf/an-iterative-regularization-method-for-total-variation-based-3dl3sp5cav.pdf

An Iterative Regularization Method for Total Variation-Based Image Restoration

The Visual Display of Quantitative Information

We present a novel approach to low-level vision problems that combines sparse coding and deep networks pre-trained with denoising auto-encoder (DA). We propose an alternative training scheme that successfully adapts DA, originally designed for unsupervised feature learning, to the tasks of image denoising and blind inpainting. Our method's performance in the image denoising task is comparable to that of KSVD which is a widely used sparse coding technique. More importantly, in blind image inpainting task, the proposed method provides solutions to some complex problems that have not been tackled before. Specifically, we can automatically remove complex patterns like superimposed text from an image, rather than simple patterns like pixels missing at random. Moreover, the proposed method does not need the information regarding the region that requires inpainting to be given a priori. Experimental results demonstrate the effectiveness of the proposed method in the tasks of image denoising and blind inpainting. We also show that our new training scheme for DA is more effective and can improve the performance of unsupervised feature learning.

/pdf/image-denoising-and-inpainting-with-deep-neural-networks-3q06y9brfo.pdf

Image Denoising and Inpainting with Deep Neural Networks

Fast Fourier Transform

An Introduction to Ray tracing

From the Publisher:
Image synthesis, or rendering, is a field of transformation: it changes geometry and physics into meaningful images. Because the most popular algorithms frequently change, it is increasingly important for researchers and implementors to have a basic understanding of the principles of image synthesis. Focusing on theory, Andrew Glassner provides a comprehensive explanation of the three core fields of study that come together to form digital image synthesis: the human visual system, digital signal processing, and the interaction of matter and light. Assuming no more than a basic background in calculus, Glassner transforms his passion and expertise into a thorough presentation of each of these disciplines, and their elegant orchestration into modern rendering techniques such as radiosity and ray tracing.

Principles of Digital Image Synthesis

Proceedings of the 21st annual conference on Computer graphics and interactive techniques

We are on the verge of creating an exciting new kind of interactive story form that will involve audiences as active participants. This book provides a solid foundation in the fundamentals of classical story structure and classical game structure and explains why it has been surprisingly difficult to bring these two activities together. With this foundation in place, the book presents several ideas for ways to move forward in this appealing quest. The author has a conversational and friendly style, making reading a pleasure.

Interactive Storytelling: Techniques for 21st Century Fiction

We introduce an adaptive, iterative technique for obtaining texture samples of arbitrary precision when synthesizing a computer-generated image. The technique is an improvement on the sum table texturing method. To motivate the technique we analyze the error properties of the sum-table method. Based on that analysis we propose using a combination of tables independently or together to obtain a better estimate, and analyze the error properties of such methods. We then propose a new technique for obtaining texture samples whose accuracy is a function of the texture and the image. As part of this technique we propose the use of an auxiliary table which contains local estimates of the texture variance. We show how the iteration of a given sample may be controlled by values from this table. We then analyze the error in this method, and present images which demonstrate the improvement.

Andrew Glassner

Papers

An Introduction to Ray tracing

Principles of Digital Image Synthesis

Proceedings of the 21st annual conference on Computer graphics and interactive techniques

Interactive Storytelling: Techniques for 21st Century Fiction

Adaptive precision in texture mapping