Exploiting the GPU power for image-based relighting and neural network.

• Other Authors: Wei, Dan.
• Format: Others
• Language: English; Chinese
• Published: 2006
• Subjects: Computer graphics; Image processing > Digital techniques; Neural networks (Computer science)
• Online Access: http://library.cuhk.edu.hk/record=b5892778; http://repository.lib.cuhk.edu.hk/en/item/cuhk-325499

Wei Dan. === Thesis submitted in: October 2005. === Thesis (M.Phil.)--Chinese University of Hong Kong, 2006. === Includes bibliographical references (leaves 93-101). === Abstracts in English and Chinese. === Chapter 1 --- Introduction --- p.1 === Chapter 1.1 --- Background --- p.1 === Chapter 1.2 --- Our applications --- p.1 === Chapter 1.3 --- Structure of the thesis --- p.2 === Chapter 2 --- The Programmable Graphics Hardware --- p.4 === Chapter 2.1 --- Introduction --- p.4 === Chapter 2.2 --- The evolution of programmable graphics hardware --- p.4 === Chapter 2.3 --- Benefit of GPU --- p.6 === Chapter 2.4 --- Architecture of programmable graphics hardware --- p.9 === Chapter 2.4.1 --- The graphics hardware pipeline --- p.9 === Chapter 2.4.2 --- Programmable graphics hardware --- p.10 === Chapter 2.5 --- Data Mapping in GPU --- p.12 === Chapter 2.6 --- Some limitations of current GPU --- p.13 === Chapter 2.7 --- Application and Related Work --- p.16 === Chapter 3 --- Image-based Relighting on GPU --- p.18 === Chapter 3.1 --- Introduction --- p.18 === Chapter 3.2 --- Image based relighting --- p.20 === Chapter 3.2.1 --- The plenoptic illumination function --- p.20 === Chapter 3.2.2 --- Sampling and Relighting --- p.21 === Chapter 3.3 --- Linear Approximation Function --- p.22 === Chapter 3.3.1 --- Spherical harmonics basis function --- p.22 === Chapter 3.3.2 --- Radial basis function --- p.23 === Chapter 3.4 --- Data Representation --- p.23 === Chapter 3.5 --- Relighting on GPU --- p.24 === Chapter 3.5.1 --- Directional light source relighting --- p.27 === Chapter 3.5.2 --- Point light source relighting --- p.28 === Chapter 3.6 --- Experiment --- p.32 === Chapter 3.6.1 --- Visual Evaluation --- p.32 === Chapter 3.6.2 --- Statistic Evaluation --- p.33 === Chapter 3.7 --- Conclusion --- p.34 === Chapter 4 --- Texture Compression on GPU --- p.40 === Chapter 4.1 --- Introduction --- p.40 === Chapter 4.2 --- The Feature of Texture Compression --- p.41 === Chapter 4.3 --- Implementation --- p.42 === Chapter 4.3.1 --- Encoding --- p.43 === Chapter 4.3.2 --- Decoding --- p.46 === Chapter 4.4 --- The Texture Compression based Relighting on GPU --- p.46 === Chapter 4.5 --- An improvement of the existing compression techniques --- p.48 === Chapter 4.6 --- Experiment Evaluation --- p.50 === Chapter 4.7 --- Conclusion --- p.51 === Chapter 5 --- Environment Relighting on GPU --- p.55 === Chapter 5.1 --- Overview --- p.55 === Chapter 5.2 --- Related Work --- p.56 === Chapter 5.3 --- Linear Approximation Algorithm --- p.58 === Chapter 5.3.1 --- Basic Architecture --- p.58 === Chapter 5.3.2 --- Relighting on SH --- p.60 === Chapter 5.3.3 --- Relighting on RBF --- p.61 === Chapter 5.3.4 --- Sampling the Environment --- p.63 === Chapter 5.4 --- Implementation on GPU --- p.64 === Chapter 5.5 --- Evaluation --- p.66 === Chapter 5.5.1 --- Visual evaluation --- p.66 === Chapter 5.5.2 --- Statistic evaluation --- p.67 === Chapter 5.6 --- Conclusion --- p.69 === Chapter 6 --- Neocognitron on GPU --- p.70 === Chapter 6.1 --- Overview --- p.70 === Chapter 6.2 --- Neocognitron --- p.72 === Chapter 6.3 --- Neocognitron on GPU --- p.75 === Chapter 6.3.1 --- Data Mapping and Connection Texture --- p.76 === Chapter 6.3.2 --- Convolution and Offset Computation --- p.77 === Chapter 6.3.3 --- Recognition Pipeline --- p.80 === Chapter 6.4 --- Experiments and Results --- p.81 === Chapter 6.4.1 --- Performance Evaluation --- p.81 === Chapter 6.4.2 --- Feature Visualization of Intermediate-Layer --- p.84 === Chapter 6.4.3 --- A Real-Time Tracking Test --- p.84 === Chapter 6.5 --- Conclusion --- p.87 === Chapter 7 --- Conclusion --- p.90 === Bibliography --- p.93