Sketch-to-photo synthesis usually faced the problem of lack of labeled data, so we propose some methods based on CycleGAN to train a model to translate sketch to photo with unpaired data. Our main contribution is a proposed Sketch-to-Skeleton-to-Image (SSI) method, which performs skeletonization on sketches to reduce variance on the sketch data. We also tried different representations of the skeleton and different models for our task. Experiment results show that the generated image quality has a negative correlation with the sparsity of the input data.

Shan, Qi. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (leaves 124-132). / Abstracts in English and Chinese. / Chapter 1 --- Overview --- p.1 / Chapter 1.1 --- Image Blur Overview --- p.1 / Chapter 1.2 --- Blur Identification in a Transparency's Perspective --- p.3 / Chapter 1.3 --- From Transparencies to Natural Image Priors --- p.7 / Chapter 1.4 --- Discussion of the Linear Motion Model --- p.9 / Chapter 1.5 --- Binary Texture Restoration and High-Order MRF Optimization --- p.9 / Chapter 2 --- A Review on Previous Work --- p.13 / Chapter 2.1 --- Spatially-Invariant Blur Recovery --- p.13 / Chapter 2.2 --- Spatially-Variant Blur Recovery --- p.16 / Chapter 2.3 --- Markov Random Field Inference --- p.18 / Chapter 3 --- Motion Blur in a Transparency's Perspective --- p.20 / Chapter 3.1 --- Analysis of Object Motion Blur --- p.20 / Chapter 3.1.1 --- 1D Object Motion Blur --- p.20 / Chapter 3.1.2 --- 2D Object Motion Blur --- p.23 / Chapter 3.2 --- Modeling 2D Object Motion Blur --- p.26 / Chapter 3.3 --- Optimization Procedure --- p.27 / Chapter 3.3.1 --- Blur Kernel Estimation --- p.29 / Chapter 3.3.2 --- Latent Binary Matte Estimation --- p.30 / Chapter 3.4 --- Generalized Transparency in Motion Blur --- p.33 / Chapter 3.4.1 --- Camera Motion Blur Estimation --- p.35 / Chapter 3.4.2 --- Implementation --- p.37 / Chapter 3.5 --- Analysis and Results --- p.38 / Chapter 3.5.1 --- Evaluation of the Kernel Initialization --- p.40 / Chapter 3.5.2 --- Evaluation of Binary Alpha Initialization --- p.40 / Chapter 3.5.3 --- Robustness to Noise --- p.41 / Chapter 3.5.4 --- Natural Image Deblurring Results --- p.41 / Chapter 3.6 --- Proofs --- p.50 / Chapter 4 --- Rotational Motion Deblurring --- p.55 / Chapter 4.1 --- Motion blur descriptor --- p.55 / Chapter 4.1.1 --- Descriptor analysis --- p.56 / Chapter 4.2 --- Optimization --- p.59 / Chapter 4.2.1 --- Parameter initialization --- p.59 / Chapter 4.2.2 --- Iterative optimization --- p.62 / Chapter 4.2.3 --- Recover the color image --- p.65 / Chapter 4.3 --- Result and analysis --- p.65 / Chapter 5 --- Image Deblurring using Natural Image Priors --- p.70 / Chapter 5.1 --- Problem Definition --- p.70 / Chapter 5.2 --- Analysis of Ringing Artifacts --- p.71 / Chapter 5.3 --- Our model --- p.74 / Chapter 5.3.1 --- Definition of the probability terms --- p.75 / Chapter 5.4 --- Optimization --- p.81 / Chapter 5.4.1 --- Optimizing L --- p.83 / Chapter 5.4.2 --- Optimizing f --- p.86 / Chapter 5.4.3 --- Optimization Details and Parameters --- p.87 / Chapter 5.5 --- Experimental Results --- p.90 / Chapter 6 --- High Order MRF and its Optimization --- p.94 / Chapter 6.1 --- The Approach --- p.95 / Chapter 6.1.1 --- Polynomial Standardization --- p.95 / Chapter 6.1.2 --- Polynomial Graph Construction --- p.97 / Chapter 6.1.3 --- Polynomial Graph Partition --- p.103 / Chapter 6.1.4 --- Multi-Label Expansion --- p.105 / Chapter 6.1.5 --- Analysis --- p.106 / Chapter 6.2 --- Experimental Results --- p.108 / Chapter 6.3 --- Summary --- p.112 / Chapter 6.4 --- Proofs --- p.112 / Chapter 7 --- Conclusion --- p.117 / Chapter 7.1 --- Solving Linear Motion Blur in a Transparency's Perspective --- p.117 / Chapter 7.2 --- Rotational Motion Deblurring --- p.119 / Chapter 7.3 --- Image Deblurring using Natural Image Priors --- p.119 / Chapter 7.4 --- Contribution --- p.121 / Chapter 7.5 --- Discussion and Open Questions --- p.121 / Bibliography --- p.124

Huang, Ting. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (p. 37-41). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Related Area --- p.2 / Chapter 1.2 --- Previous Work --- p.4 / Chapter 1.3 --- Proposed Framework --- p.7 / Chapter 1.4 --- Overview --- p.8 / Chapter 2 --- Markov Random Fields and Optimization Schemes --- p.9 / Chapter 2.1 --- MRF Model --- p.10 / Chapter 2.1.1 --- MAP Understanding --- p.11 / Chapter 2.2 --- Belief Propagation Optimization Scheme --- p.14 / Chapter 2.2.1 --- Max-Product BP on MRFs --- p.14 / Chapter 2.2.2 --- Sum-Product BP on MRFs --- p.15 / Chapter 3 --- Our Formulation --- p.17 / Chapter 3.1 --- An MRF Model --- p.18 / Chapter 3.2 --- Coarse-to-Fine Optimization by BP --- p.21 / Chapter 3.2.1 --- Coarse-Level Belief Propagation --- p.23 / Chapter 3.2.2 --- Fine-Level Belief Propagation --- p.24 / Chapter 3.2.3 --- Performance Enhancement --- p.25 / Chapter 4 --- Experiments --- p.27 / Chapter 4.1 --- Comparison --- p.27 / Chapter 4.2 --- Failure Case --- p.32 / Chapter 5 --- Conclusion --- p.35 / Bibliography --- p.37

Thresholding algorithms and filters are of great importance when utilizing OCR to extract information from text documents such as invoices. Invoice documents vary greatly and since the performance of image processing methods when applied to those documents will vary accordingly, selecting appropriate methods is critical if a high recognition rate is to be obtained. This paper aims to determine if a document recognition system that automatically selects optimal processing methods, based on the characteristics of input images, will yield a higher recognition rate than what can be achieved by a manual choice. Such a recognition system, including a learning framework for selecting optimal thresholding algorithms and filters, was developed and evaluated. It was established that an automatic selection will ensure a high recognition rate when applied to a set of arbitrary invoice images by successfully adapting and avoiding the methods that yield poor recognition rates.

In stochastic representation an image is a sample function of an array of random variables which is called a random field. For characterizing an ensemble of images, we choose an autoregressive model as our image model. An image model often applies to image processing such as image data compression and image restoration. Therefore the validity of the image model affect it¡¦s performance of image processing. The output of the AR model depends on its parameters ¡V system transition matrix and generating noise. Hence the validity of this model is related to these two parameters. How to seek the standard of the validity of the image model is a problem. We exploit performance of image model¡¦s application ¡V image restoration - to find a method of determining the validity of the image model. In our paper we find a relation between image restoration performance and image model¡¦s parameters by the Kalman filtering equations. An image model with lower generating noise power and system transition matrix is better for image restoration and is considered a good image model. In the analysis of the parameters of the image model, we can meet the requirements of the parameters by image segmentation method, residual image method and normalized image method. In addition it also helps us understand the Kalman filter much more and know how to find the solution of similar problems.

<p>The target of this master thesis is to evaluate the Image Warping technique and propose a possible design for an implementation in FPGA. The Image Warping is widely used in the image processing for image correction and rectification. A DSP is a usual choice for implantation of the image processing algorithms, but to decrease a cost of the target system it was proposed to use an FPGA for implementation.</p><p>In this work a different Image Warping methods was evaluated in terms of performance, produced image quality, complexity and design size. Also, considering that it is not only Image Warping algorithm which will be implemented on the target system, it was important to estimate a possible memory bandwidth used by the proposed design. The evaluation was done by implemented a C-model of the proposed design with a finite datapath to simulate hardware implementation as close as possible.</p>

Title from title page of PDF (University of Missouri--St. Louis, viewed February 8, 2010). Includes bibliographical references (p. 69-73).

Progress in multimedia technologies has resulted in a plethora of services and devices that capture, compress, transmit and display audiovisual stimuli. Humans -- the ultimate receivers of such stimuli -- now have access to visual entertainment at their homes, their workplaces as well as on mobile devices. With increasing visual signals being received by human observers, in the face of degradations that occur to due the capture, compression and transmission processes, an important aspect of the quality of experience of such stimuli is the \emph{perceived visual quality}. This dissertation focuses on algorithm development for assessing such visual quality of natural images, without need for the `pristine' reference image, i.e., we develop computational models for no-reference image quality assessment (NR IQA). Our NR IQA model stems from the theory that natural images have certain statistical properties that are violated in the presence of degradations, and quantifying such deviations from \emph{naturalness} leads to a blind estimate of quality. The proposed modular and easily extensible framework is distortion-agnostic, in that it does not need to have knowledge of the distortion afflicting the image (contrary to most present-day NR IQA algorithms) and is not only capable of quality assessment with high correlation with human perception, but also is capable of identifying the distortion afflicting the image. This additional distortion-identification, coupled with blind quality assessment leads to a framework that allows for blind general-purpose image repair, which is the second major contribution of this dissertation. The blind general-purpose image repair framework, and its exemplar algorithm described here stem from a revolutionary perspective on image repair, where the framework does not simply attempt to ameliorate the distortion in the image, but to ameliorate the distortion, so that visual quality at the output is maximized. Lastly, this dissertation describes a large-scale human subjective study that was conducted at UT to assess human behavior and opinion on visual quality of videos when viewed on mobile devices. The study lead to a database of 200 distorted videos, which incorporates previously studied distortions such as compression and wireless packet-loss, and also dynamically varying distortions that change as a function of time, such as frame-freezes and temporally varying compression rates. This study -- the first of its kind -- involved over 50 human subjects and resulted in 5,300 summary subjective scores and time-sampled subjective traces of quality for multiple displays. The last part of this dissertation analyzes human behavior and opinion on time-varying video quality, opening up an extremely interesting and relevant field for future research in the area of quality assessment and human behavior. / text

Thesis (Ph.D.)--George Mason University, 2008. / Vita: p. 133. Thesis directors: Zoran Duric, Dana Richards. Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Computer Science. Title from PDF t.p. (viewed Mar. 17, 2009). Includes bibliographical references (p. 127-132). Also issued in print.

Dissertation (Ph.D.)--Auburn University, 2006. / Abstract. Vita. Includes bibliographic references (p.157-171).