Browsing by Issue Date, starting with "2019-02-05"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
- Structure tensor-based depth estimation from light field imagesPublication . Lourenço, Rui Miguel Leonel; Assunção, Pedro António Amado de; Távora, Luís Miguel de Oliveira Pegado de Noronha eThis thesis presents a novel framework for depth estimation from light eld images based on the use of the structure tensor. A study of prior knowledge introduces general concepts of depth estimation from light eld images. This is followed by a study of the state-of-the art, including a discussion of several distinct depth estimation methods and an explanation of the structure tensor and how it has been used to acquire depth estimation from a light eld image. The framework developed improves on two limitations of traditional structure tensor derived depth maps. In traditional approaches, foreground objects present enlarged boundaries in the estimated disparity map. This is known as silhouette enlargement. The proposed method for silhouette enhancement uses edge detection algorithms on both the epipolar plane images and their corresponding structure tensor-based disparity estimation and analyses the di erence in the position of these di erent edges to establish a map of the erroneous regions. These regions can be inpainted with values from the correct region. Additionally, a method was developed to enhance edge information by linking edge segments. Structure tensor-based methods produce results with some noise. This increases the di culty of using the resulting depth maps to estimate the orientation of scenic surfaces, since the di erence between the disparity of adjacent pixels often does not correlate with the real orientation of the scenic structure. To address this limitation, a seed growing approach was adopted, detecting and tting image planes in a least squares sense, and using the estimated planes to calculate the depth for the corresponding planar region. The full framework provides signi cant improvements on previous structure tensorbased methods. When compared with other state-of-the-art methods, it proves competitive in both mean square error and mean angle error, with no single method proving superior in every metric.
- Improved image rendering for focused plenoptic cameras with extended depth-of-fieldPublication . Filipe, José Nunes dos Santos; Távora, Luís Miguel Oliveira Pegado de Noronha e; Faria, Sérgio Manuel Maciel deThis dissertation presents a research work on rendering images from light elds captured with a focused plenoptic camera with extended depth of eld. A basic overview of the 7 dimensional plenoptic function is rst given, followed by a description of the Two-Plane Parametrisation. Some of the various methods used for sampling the plenoptic functions are then described, namely those equivalent to acquisition functions implemented by the camera gantry, the unfocused plenoptic camera and the focused plenoptic camera. State-of-the-art image rendering algorithms have also been studied both for focused and unfocused plenoptic cameras. A comprehensive study of the behaviour of focus metrics when applied to images rendered form a focused plenoptic camera is presented, including 34 of the most widely used metrics in the literature. Due to high frequency artefacts, caused by the rendering process, it was found that the currently available focus metrics yield in ated values for this kind of images, leading to misindication, where worse-focused images have better focus measures. Subjective tests were carried out, in order to corroborate these results. Then, methods for minimizing the rendering artefacts are proposed. An algorithm for choosing the maximum patch size for each micro-image was designed, in order to minimize the distortions caused by the vignetting e ect of the micro-lens. Then an inpainting algorithm, based on anisotropic di usion inpainting, is used to minimize the remaining artefacts present in the borders between adjacent micro-images. Finally, a method to deal with the redundant information generated by a plenoptic camera with extended depth of eld is presented. Three di erent views of the same scene are rendered, with the three di erent types of lenses. Then, it is proven that making any linear combination of the images always results in worse focus than selecting the better focused one. Thus, a multi-focus image fusion algorithm is proposed to merge the three images captured by a extended depth-of- eld camera into a single one, which presents higher focus level than any of the three individual images.
- Fast video encoding based on random forestsPublication . Tahir, Muhammad; Taj, Imtiaz A.; Assuncao, Pedro A. A.; Asif, MuhammadMachine learning approaches have been increasingly used to reduce the high computational complexity of high-efficiency video coding (HEVC), as this is a major limiting factor for real-time implementations, due to the decision process required to find optimal coding modes and partition sizes for the quad-tree data structures defined by the standard. This paper proposes a systematic approach to reduce the computational complexity of HEVC based on an ensemble of online and offline Random Forests classifiers. A reduced set of features for training the Random Forests classifier is proposed, based on the rankings obtained from information gain and a wrapper-based approach. The best model parameters are also obtained through a consistent and generalizable method. The proposed Random Forests classifier is used to model the coding unit and transform unit-splitting decision and the SKIP-mode prediction, as binary classification problems, taking advantage from the combination of online and offline approaches, which adapts better to the dynamic characteristics of video content. Experimental results show that, on average, the proposed approach reduces the computational complexity of HEVC by 62.64% for the random access (RA) profile and 54.57% for the low-delay (LD) main profile, with an increase in BD-Rate of 2.58% for RA and 2.97% for LD, respectively. These results outperform the previous works also using ensemble classifiers for the same purpose.