3dinfogan: 3d Models’ Reconstruction In Infogans

In computer vision and computer graphics, 3D reconstruction is the process of capturing real objects’ shapes and appearances. 3D models always can be constructed by active methods which use high-quality scanner equipment, or passive methods that learn from the dataset. However, both of these two methods only aimed to construct the 3D models, without showing what element affects the generation of 3D models. Therefore, the goal of this research is to apply deep learning to automatically generating 3D models, and finding the latent variables which affect the reconstructing process. The existing research GANs can be trained in little data with two networks called Generator and Discriminator, respectively. Generator can produce synthetic data, and Discriminator can discriminate between the generator’s output and real data. The existing research shows that InFoGAN can maximize the mutual information between latent variables and observation. In our approach, we will generate the 3D models based on InFoGAN and design two constraints, shape-constraint and parameters-constraint, respectively. Shape-constraint utilizes the data augmentation method to limit the synthetic data generated in the models’ profiles. At the same time, we also try to employ parameters-constraint to find the 3D models’ relationship corresponding to the latent variables. Furthermore, our approach will be a challenge in the architecture of generating 3D models built on InFoGAN. Finally, in the process of generation, we might discover the contribution of the latent variables influencing the 3D models to the whole network.

Hippocampus segmentation and classification for dementia analysis using pre-trained neural network models

The diagnostic and clinical overlap of early mild cognitive impairment (EMCI), mild cognitive impairment (MCI), late mild cognitive impairment (LMCI) and Alzheimer disease (AD) is a vital oncological issue in dementia disorder. This study is designed to examine Whole brain (WB), grey matter (GM) and Hippocampus (HC) morphological variation and identify the prominent biomarkers in MR brain images of demented subjects to understand the severity progression. Curve evolution based on shape constraint is carried out to segment the complex brain structure such as HC and GM. Pre-trained models are used to observe the severity variation in these regions. This work is evaluated on ADNI database. The outcome of the proposed work shows that curve evolution method could segment HC and GM regions with better correlation. Pre-trained models are able to show significant severity difference among WB, GM and HC regions for the considered classes. Further, prominent variation is observed between AD vs. EMCI, AD vs. MCI and AD vs. LMCI in the whole brain, GM and HC. It is concluded that AlexNet model for HC region result in better classification for AD vs. EMCI, AD vs. MCI and AD vs. LMCI with an accuracy of 93, 78.3 and 91% respectively.

Target Design For Lida-Based ICP Pose Estimation For Space Vision Tasks

The goal of this thesis is to develop a methodology for designing 3D target shapes for accurate LIDAR pose estimation. Scanned from a range of views, this shape can be attached to the surface of a spacecraft and deliver accurate pose scanned. It would act as an LIDAR- based analogue to fiducial markers placed on the surface and viewed by CCD camera(s). Continuum Shape Constraint Analysis (CSCA) which assesses shapes for pose estimation and measures the performance of the Iterative Closest Point (ICP) Algorithm is used as a shape design tool. CSCA directly assesses the sensitivity of pose error to variation in viewing direction. Three of the CSCA measures, Noise Amplification Index, Minimal Eigen-value and Expectivity Index, were compared, and Expectivity Index was shown to be the best index to use as shape design tool. Using CSCA and numerical simulations, a Cuboctahedron was shown to be an optimal shape which delivers an accurate pose when viewed from all angles and the nitial pose guess is close to the true poses. Separate from Constraint Analysis, the problem of shape ambiguity was addressed using numerical tools. The Cuboctahedron was modified in order to resolve shape ambiguity - the tendency of the ICP algorithm to converge with low registration error on a pose configuration geometrically identical, but actually different from a “true pose”. The numerical characteristics of geometrical ambiguity were studied, and a heuristic design methodology to reduce shape ambiguity was developed and is presented in this thesis. A Reduced Ambiguity Cuboctahedron is the resultant shape that delivers an accurate pose from all views and does not suffer from shape ambiguity. The shapes were subjected to simulation and experimental validation. They were manufactured using 3D Rapid Prototyper, and a NEPTEC Design Group TriDAR Scanner was used to obtain experimental data for three shapes: the Tetrahedron, Cuboctahedron, and reduced Ambiguity Cuboctahedron. The Tetrahedron, which has poorly constrained views, was included in the testing process as a comparison shape. The simulation and experimental results were congruent, and validated the design methodology and the designed shapes.

Target Design For Lida-Based ICP Pose Estimation For Space Vision Tasks

The goal of this thesis is to develop a methodology for designing 3D target shapes for accurate LIDAR pose estimation. Scanned from a range of views, this shape can be attached to the surface of a spacecraft and deliver accurate pose scanned. It would act as an LIDAR- based analogue to fiducial markers placed on the surface and viewed by CCD camera(s). Continuum Shape Constraint Analysis (CSCA) which assesses shapes for pose estimation and measures the performance of the Iterative Closest Point (ICP) Algorithm is used as a shape design tool. CSCA directly assesses the sensitivity of pose error to variation in viewing direction. Three of the CSCA measures, Noise Amplification Index, Minimal Eigen-value and Expectivity Index, were compared, and Expectivity Index was shown to be the best index to use as shape design tool. Using CSCA and numerical simulations, a Cuboctahedron was shown to be an optimal shape which delivers an accurate pose when viewed from all angles and the nitial pose guess is close to the true poses. Separate from Constraint Analysis, the problem of shape ambiguity was addressed using numerical tools. The Cuboctahedron was modified in order to resolve shape ambiguity - the tendency of the ICP algorithm to converge with low registration error on a pose configuration geometrically identical, but actually different from a “true pose”. The numerical characteristics of geometrical ambiguity were studied, and a heuristic design methodology to reduce shape ambiguity was developed and is presented in this thesis. A Reduced Ambiguity Cuboctahedron is the resultant shape that delivers an accurate pose from all views and does not suffer from shape ambiguity. The shapes were subjected to simulation and experimental validation. They were manufactured using 3D Rapid Prototyper, and a NEPTEC Design Group TriDAR Scanner was used to obtain experimental data for three shapes: the Tetrahedron, Cuboctahedron, and reduced Ambiguity Cuboctahedron. The Tetrahedron, which has poorly constrained views, was included in the testing process as a comparison shape. The simulation and experimental results were congruent, and validated the design methodology and the designed shapes.

Large Deviations for a Class of Multivariate Heavy-Tailed Risk Processes Used in Insurance and Finance

Modern risk modelling approaches deal with vectors of multiple components. The components could be, for example, returns of financial instruments or losses within an insurance portfolio concerning different lines of business. One of the main problems is to decide if there is any type of dependence between the components of the vector and, if so, what type of dependence structure should be used for accurate modelling. We study a class of heavy-tailed multivariate random vectors under a non-parametric shape constraint on the tail decay rate. This class contains, for instance, elliptical distributions whose tail is in the intermediate heavy-tailed regime, which includes Weibull and lognormal type tails. The study derives asymptotic approximations for tail events of random walks. Consequently, a full large deviations principle is obtained under, essentially, minimal assumptions. As an application, an optimisation method for a large class of Quota Share (QS) risk sharing schemes used in insurance and finance is obtained.