We consider the problem of localizing objects in weakly labeled images/videos. An image/video (e.g., Flickr image and YouTube video) is weakly labeled if it is associated with a tag describing the main object present in the image/video. It is weakly labeled because the tag only indicates the presence/absence of the object, but does not provide the detailed spatial location of the object. Given an image/video with an object tag, our goal is to localize the object in it. In this thesis, we propose two novel techniques to handle this challenging problem. First, we build a video-specific object appearance model and then incorporate temporal consistency information to localize the object. Second, we make use of existing detectors of some other object classes (which we call "familiar objects") to build the appearance model of the unseen object class (i.e., the object of interest). Experimental results show the effectiveness of the proposed methods. / October 2016

This thesis addresses the problem of automatic detection and identification of blemishes in digital images of potatoes. Potatoes are an important food crop, with clear unblemished skin being the main factor affecting consumer preference. Potatoes with defects, diseases and blemishes caused by otherwise benign (to human) skin infections, are strongly avoided by consumers. Most potatoes are sorted into dfferent grades by hand, with inevitable mistakes and losses. The currently deployed computer vision systems for sorting potatoes require manual training and have limited accuracy and high unit costs. A further limitation of typical machine vision systems is that the set of image features for pattern recognition has to be designed by the system engineer to work with a specific configuration of produce, imaging system and operating conditions. Such systems typically do not generalise well to other configurations, where the required image features may well differ from those used to design the original system. The objective of the research presented in this thesis is to introduce an automatic method for detecting and identifying blemishes in digital images of potatoes, where the presented solution involves classifying individual pixels. A human expert is required to mark up areas of blemishes and non-blemishes in a set of training images. For blemish detection, each pixel is classified as either blemish or non-blemish. For blemish identification, each pixel is classified according to a number of pre-determined blemish categories. After training, the system should be able to classify individual pixels in new images of previously unseen potatoes with high accuracy. After segmenting the potato from the image background, a very large set of candidate features, based on statistical information relating to the colour and texture of the region surrounding a given pixel, is first extracted. The features include statistical summaries of the whole potato and local regions centred on each pixel as well as the data of the pixel itself. Then an adaptive boosting algorithm (AdaBoost) is used to automatically select the best features for discriminating between blemishes and non-blemishes. The AdaBoost algorithm (Freund and Schapire, 1999) is used to build a classifier, which combines results from so-called "weak" classifiers, each constructed using one of the candidate features, into one "strong" classifier that performs better than any of the weak classifiers alone. With this approach, different features can be selected for different potato varieties, while also handling the natural variation in fresh produce due to different seasons, lighting conditions, etc. For blemish detection, the classifier was trained using a subset of pixels which had been marked as blemish or non-blemish. Tests were done with the full set of features, "lesion experiments" were carried out to explore the impact of removing specific feature types, and experiments were also carried out on methods of speeding up classification both by restricting the number of weak classifiers and restricting the numbers of unique candidate features which can be used to produce weak classifiers. The results were highly accurate with visible examples of disagreement between classifier output and markup being caused by human inaccuracies in the markup rather than classifier inaccuracy. For blemish identification, a set of classifiers were trained on subsets of pixels marked as each blemish class against a subset of pixels drawn from all other classes. For classification, each pixel was tested with all classifiers and assigned to the classifier which returned the highest confidence of a positive result. Experiments were again performed with methods of speeding up classification as well as lesion experiments. Finally, to demonstrate how the system would work in an industrial context, the classification results were summarised for each potato, providing a high overall accuracy in detecting the presence or absence of significant blemish coverage for each blemish type.

No description available.

This thesis introduces the problem of learning the representation and the classification of the perceived sense of movement, defined as dynamism in static visuals. To solve the said problem, we study the definition, degree, and real-world implications of dynamism within the field of consumer psychology. We employ Deep Convolutional Neural Networks (DCNN) as a method to learn and predict dynamism in images. The novelty of the task, lead us to collect a dataset which we synthetically augmented for spatial invariance, using image processing techniques. We study the methods of transfer learning to transfer knowledge from another domain, as the size of our dataset was deemed to be inadequate. Our dataset is trained across different network architectures, and transfer learning techniques to find an optimal method for the task at hand. To show a real-world application of our work, we observe the correlation between the two visual stimuli, dynamism and emotions.

In both RMCS calibration and ACS calibration, the corresponding efficiency and robustness are tested by simulation and real experiments. In the real experiment of ACS calibration, the intrinsic and extrinsic parameters of the ACS are obtained simultaneously by our calibration procedure using the same image sequences, no extra data capturing step is required. The corresponding trajectory is recovered and illustrated using the calibration results of the ACS. Since the estimated translations of different cameras in an MCS may scaled by different scale factors, scale factor estimation algorithms are proposed for non-overlapping view RMCS calibration and ACS calibration respectively. To our knowledge, we are the first to study the calibration of ACS. / In this thesis, we focus on developing robust methods for the MCS calibration problems. In particular, we make two contributions. Firstly, we developed a novel extrinsic calibration method for the non-overlapping view Rigid Multiple Camera System (RMCS) using the kinematic information of the RMCS. The input are only the images captured when the non-overlapping RMCS is moved in an environment with enough static feature points. This assumption is true in many vision tasks such as SFM (Structure from Motion), SLAM (Simultaneous Localization and Map). The output is the extrinsic parameters of the cameras of the RMCS. / Multiple Camera Systems (MCS) have been widely applied in many vision applications and attracted much attention recently. Both intrinsic and extrinsic parameters of an MCS are needed to be calibrated before it is used. / Secondly, we proposed to solve the calibration of a particular model of non-rigid Multiple Camera System, namely, Articulated Camera System (ACS). In an ACS, the cameras are fixed on articulated arms with joints, the relative pose between them may change. Two ACS calibration methods are proposed. In the first approach, we assume the cameras have overlapping views. It uses the feature correspondences between the cameras in the ACS. In the second approach, we assume the cameras have no overlapping view. It requires only the ego-motion information of the cameras and can be used for the calibration of the non-overlapping view ACS. In both methods, the ACS is assumed to have performed general transformations in a static environment. / Chen, Junzhou. / Adviser: Kin Hong Wong. / Source: Dissertation Abstracts International, Volume: 70-06, Section: B, page: 3594. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (leaves 102-110). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

by Lam Shu Sun. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1993. / Includes bibliographical references (leaves [91]-94). / Chapter Chapter1 --- Introduction: Binocular Depth Visual Perception of Human --- p.1 / Chapter 1.1 --- Introduction --- p.1 / Chapter 1.2 --- The visual pathway --- p.2 / Chapter 1.3 --- The retina --- p.3 / Chapter 1.4 --- The ganglion cells --- p.5 / Chapter 1.5 --- The lateral geniculate nucleus --- p.7 / Chapter 1.6 --- The visual cortex --- p.8 / Chapter 1.6.1 --- The cortical cells --- p.8 / Chapter 1.6.2 --- The organization of the visual cortex --- p.9 / Chapter 1.7 --- Stereopsis --- p.11 / Chapter 1.7.1 --- Corresponding retinal points --- p.12 / Chapter 1.7.2 --- Binocular fusion --- p.14 / Chapter 1.7.3 --- The binocular depth cells --- p.14 / Chapter 1.8 --- Conclusion of chapter 1 --- p.15 / Chapter Chapter2 --- Computational Stereo Vision --- p.15 / Chapter 2.1 --- Stereo image geometry --- p.16 / Chapter 2.1.1 --- The crossed-looking geometry --- p.17 / Chapter 2.1.2 --- The parallel optical axes geometry --- p.19 / Chapter 2.2 --- The false targets problem --- p.20 / Chapter 2.3 --- Feature selection --- p.21 / Chapter 2.3.1 --- Zero-crossing method --- p.21 / Chapter 2.3.2 --- A network model for ganglion cell --- p.24 / Chapter 2.4 --- The constraints of matching --- p.28 / Chapter 2.5 --- Correspondence techniques --- p.29 / Chapter 2.6 --- Conclusion of chapter 2 --- p.29 / Chapter Chapter3 --- A Hypercolumn Based Stereo Vision Model --- p.30 / Chapter 3.1 --- A visual model for stereo vision --- p.30 / Chapter 3.2 --- The model of PSVM (A Computerized Visual Model) --- p.32 / Chapter 3.3 --- Local orientated line extraction (Stage 1 of PSVM) --- p.34 / Chapter 3.3.1 --- Orientated line detection network --- p.35 / Chapter 3.3.2 --- On-type orientated lines and off-type orientated lines --- p.37 / Chapter 3.4 --- Local line matching (Stage 2 of PSVM) --- p.38 / Chapter 3.4.1 --- Structure of hypercolumn in PSVM --- p.39 / Chapter 3.4.2 --- Line length discrimination model (Part of stage 2 of PSVM) --- p.41 / Chapter 3.4.3 --- Orientation-length detector --- p.42 / Chapter 3.4.4 --- Line length selection --- p.45 / Chapter 3.4.5 --- The matching model --- p.46 / Chapter 3.4.6 --- Fusional area in PSVM --- p.48 / Chapter 3.4.7 --- Matching mechanism --- p.49 / Chapter 3.4.8 --- Disparity detection --- p.50 / Chapter 3.5 --- Disparity integrations (Stage 3 of PSVM) --- p.53 / Chapter 3.5.1 --- The voter network --- p.54 / Chapter 3.5.2 --- The redistributor network --- p.55 / Chapter 3.6 --- Conculsion of chpater 3 --- p.57 / Chapter Chapter4 --- Implementation and Analysis --- p.58 / Chapter 4.1 --- The imaging geometry of PSVM --- p.58 / Chapter 4.2 --- Input --- p.59 / Chapter 4.3 --- The hypercolumn construction --- p.59 / Chapter 4.4 --- Analysis of matching mechanism in PSVM --- p.59 / Chapter 4.4.1 --- Fusional condition --- p.61 / Chapter 4.4.2 --- Disparity detection --- p.61 / Chapter 4.5. --- Matching rules in PSVM --- p.63 / Chapter 4.5.1 --- The ordering constraint --- p.63 / Chapter 4.5.2 --- The uniqueness constraint --- p.64 / Chapter 4.5.3 --- The figural continuity constraint --- p.64 / Chapter 4.5.4 --- The smoothness assumption --- p.65 / Chapter 4.6. --- Use multi-lengths of oriented line to solve the occlusion problem --- p.66 / Chapter 4.7 --- Performance of PSVM --- p.67 / Chapter 4.7.1 --- Artificial scene --- p.67 / Chapter 4.7.2 --- Natural images --- p.71 / Chapter 4.8 --- Discussion --- p.83 / Chapter 4.9 --- Overall conclusion --- p.83 / Appendix: Illustration example --- p.85 / References --- p.91

Xiong, Wei. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (leaves 56-61). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Background Study --- p.6 / Chapter 2.1 --- Stereo matching --- p.6 / Chapter 2.2 --- Digital image matting --- p.8 / Chapter 2.3 --- Expectation Maximization --- p.9 / Chapter 3 --- Model Definition --- p.12 / Chapter 4 --- Initialization --- p.20 / Chapter 4.1 --- Initializing disparity --- p.20 / Chapter 4.2 --- Initializing alpha matte --- p.24 / Chapter 5 --- Optimization --- p.26 / Chapter 5.1 --- Expectation Step --- p.27 / Chapter 5.1.1 --- "Computing E((Pp(df = d1̐ưجθ(n),U))" --- p.28 / Chapter 5.1.2 --- "Computing E((Pp(db = d2̐ưجθ(n),U))" --- p.29 / Chapter 5.2 --- Maximization Step --- p.31 / Chapter 5.2.1 --- "Optimize α, given {F, B} fixed" --- p.34 / Chapter 5.2.2 --- "Optimize {F, B}, given α fixed" --- p.37 / Chapter 5.3 --- Computing Final Disparities --- p.40 / Chapter 6 --- Experiment Results --- p.42 / Chapter 7 --- Conclusion --- p.54 / Bibliography --- p.56

by Siu-Yuk Yeung. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1999. / Includes bibliographical references (leaves 96-102). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.5 / Chapter 1.1 --- Shape recovery techniques --- p.5 / Chapter 1.2 --- Shape from Shading algorithms --- p.8 / Chapter 1.2.1 --- Some developments on surface reflection --- p.9 / Chapter 1.2.2 --- Some developments on computing methods --- p.11 / Chapter 1.2.3 --- Some developments on light source model --- p.12 / Chapter 1.3 --- Proposed algorithms in this thesis --- p.13 / Chapter 1.4 --- Thesis outline --- p.14 / Chapter 2 --- Camera and surface reflectance models for SFS --- p.15 / Chapter 2.1 --- Camera models for SFS --- p.16 / Chapter 2.1.1 --- Pinhole camera model and perspective projection --- p.17 / Chapter 2.1.2 --- Approximations of perspective projection --- p.20 / Chapter 2.2 --- Surface reflectance models for SFS --- p.22 / Chapter 2.2.1 --- Lambertian surface model --- p.23 / Chapter 2.2.2 --- Bidirectional Reflectance Distribuction Function --- p.23 / Chapter 2.3 --- Summary --- p.25 / Chapter 3 --- Review of some related SFS algorithms --- p.26 / Chapter 3.1 --- The SFS algorithm proposed by Bichsel and Pentland --- p.27 / Chapter 3.1.1 --- Determine surface height with a minimum downhill principle --- p.28 / Chapter 3.1.2 --- Implementation on a discrete grid --- p.30 / Chapter 3.2 --- The SFS algorithm proposed by Kimmel and Bruckstein --- p.31 / Chapter 3.2.1 --- Level set propagation --- p.32 / Chapter 3.2.2 --- Problem formulation --- p.33 / Chapter 3.2.3 --- Equal height contour propagation using level set method --- p.35 / Chapter 3.3 --- Summary --- p.36 / Chapter 4 --- Multiple extended light source models for SFS --- p.38 / Chapter 4.1 --- Three extended light source models for SFS --- p.40 / Chapter 4.1.1 --- Rectangular light source model --- p.40 / Chapter 4.1.2 --- Spherical light source model --- p.43 / Chapter 4.1.3 --- Cylindrical light source model --- p.48 / Chapter 4.2 --- SFS for an extended light source --- p.53 / Chapter 4.3 --- Multiple extended light source model --- p.53 / Chapter 4.4 --- Simulation and experiment result --- p.54 / Chapter 4.5 --- Error Analysis --- p.55 / Chapter 4.5.1 --- Descriptions of the error --- p.55 / Chapter 4.5.2 --- Errors for different light models --- p.55 / Chapter 4.6 --- Summary --- p.57 / Chapter 5 --- Global SFS for an endoscope image --- p.70 / Chapter 5.1 --- Introduction --- p.71 / Chapter 5.2 --- Local SFS algorithm for endoscope image --- p.73 / Chapter 5.2.1 --- Imaging system and brightness formulation --- p.74 / Chapter 5.2.2 --- Equal distance contour propagation and shape reconstruc- tion --- p.75 / Chapter 5.3 --- Global SFS algorithm for endoscope image --- p.76 / Chapter 5.3.1 --- A global shape from shading algorithm for a parallel light --- p.77 / Chapter 5.3.2 --- The relationship between depth map and distance map --- p.78 / Chapter 5.3.3 --- A global shape from shading algorithm for endoscope image --- p.78 / Chapter 5.4 --- Simulations and experiments results --- p.83 / Chapter 5.5 --- Summary --- p.86 / Chapter 6 --- Summary and conclusion --- p.87 / Chapter 6.1 --- Problems tackled in this thesis --- p.87 / Chapter 6.2 --- Discussion on future developments --- p.88

Stereo vision from short-baseline image pairs is one of the most active research fields in computer vision. The estimation of dense disparity maps from stereo image pairs is still a challenging task and there is further space for improving accuracy, minimizing the computational cost and handling more efficiently outliers, low-textured areas, repeated textures, disparity discontinuities and light variations. This PhD thesis presents two novel methodologies relating to stereo vision from short-baseline image pairs: I. The first methodology combines three different cost metrics, defined using colour, the CENSUS transform and SIFT (Scale Invariant Feature Transform) coefficients. The selected cost metrics are aggregated based on an adaptive weights approach, in order to calculate their corresponding cost volumes. The resulting cost volumes are merged into a combined one, following a novel two-phase strategy, which is further refined by exploiting semi-global optimization. A mean-shift segmentation-driven approach is exploited to deal with outliers in the disparity maps. Additionally, low-textured areas are handled using disparity histogram analysis, which allows for reliable disparity plane fitting on these areas. II. The second methodology relies on content-based guided image filtering and weighted semi-global optimization. Initially, the approach uses a pixel-based cost term that combines gradient, Gabor-Feature and colour information. The pixel-based matching costs are filtered by applying guided image filtering, which relies on support windows of two different sizes. In this way, two filtered costs are estimated for each pixel. Among the two filtered costs, the one that will be finally assigned to each pixel, depends on the local image content around this pixel. The filtered cost volume is further refined by exploiting weighted semi-global optimization, which improves the disparity accuracy. The handling of the occluded areas is enhanced by incorporating a straightforward and time efficient scheme. The evaluation results show that both methodologies are very accurate, since they handle efficiently low-textured/occluded areas and disparity discontinuities. Additionally, the second approach has very low computational complexity. Except for the aforementioned two methodologies that use as input short-baseline image pairs, this PhD thesis presents a novel methodology for generating 3D point clouds of good accuracy from wide-baseline stereo pairs.

This PhD is a practice-based study of how the computer functions in art practice, which takes on the notion of a fine art computing “medium”. Current research, while sometimes referencing the computer as a potential art medium, mostly defines it non-explicitly as a type of “hybrid” media device or some sort of “multimedia” machine. These terms leave the existence of a specific computing medium in art practice undefined and have historically led the analysis of artworks that employ computers to rely on critical frameworks that were either developed for earlier physical media, or have no structural similarities to computers. Such approaches can fail to examine unique ontological issues that arise - especially at a structural level - when using a computer to produce art. To achieve a formal description of a hitherto loosely defined (or non-defined) art medium, the research employs a range of critical and theoretical material from fields outside art practice, chiefly among them Alan Turing’s definition of a "a(utomatic)-machine", (nowadays called a “Turing machine”) from his 1936 paper "On Computable Numbers, with an Application to the Entscheidungsproblem". Turing described a machine which can “simulate” any other computing machine including all modern computers. His machine is here used to propose a ‘Turing-complete medium’ of art, of which every computer is a computationally equivalent member. Using this perspective/definition, the research undertook an investigation of a ‘Turing-complete medium’ by developing creative practice in the form of individual works that explored specific aspects of computing systems. The research then engaged in a written analysis of the practice, again employing the concept of a ‘Turing-complete medium’, working towards the development of medium-specific critique of any art made with any computer. In foregrounding the nature and functions of computing machines, the research explores how these elements can be made intrinsic to our interpretations of computer-based art while also being aware of the limitations of medium-specific critique as exposed within the modernist tradition.