Procedural Tree Generation: Compressing 3D tree for faster rendering

Trees are essential components of both real and digital environments. Therefore, it is important to have 3D models of trees that are of high quality and computationally efficient. One way to achieve this is by compressing a high-quality model using billboard rendering, which involves partitioning the tree into multiple planes to produce a...

Compressing code generation language models on CPUs: Using Group Lasso pruning and post-training quantization

Code generation models have become more popular recently, due to the fact that they assist developers in writing code in a more productive manner. While these large models deliver impressive performance, they require significant computational resources and memory, making them difficult to deploy and expensive to train. Additionally, their large...

CodeGPT on XTC: Compressing a CodeGPT Model Using Hybrid Layer Reduction and Extreme Quantisation through Knowledge Distillation

Large language models are powerful because of their state-of-the-art language processing abilities. But, they come at the cost of being extremely resource-intensive, and are steadily growing in size. As a result, compressing such models for resource- constrained devices is an active and promising re- search area. In spite of their current...

Exploring the potential of wavelets: In the field of image processing

This research consists of two applications of image processing, namely, image compression and image denoising. Image compression aims to reduce the size of an image without losing too many features. This is often used to store a large number of images such as fingerprints. Denoising is a technique for removing noise from an image while...

Image-based Video Search Engine: Data Compression and Nearest Neighbour Search

One of the main problems with Instance-level Image Retrieval in video data is that for longer query videos or large amount of image queries, comparing all of the query images to every extracted frame is time-inefficient. This thesis aims to solve this problem by implementing Nearest Neighbour Search (NNS) algorithms and data compression methods,...

Minimizing bandwidth utilization for streaming noisy Monte-Carlo renders: For the 2018 individual research pilot

<br/>This study focuses on the question how the bandwidth utilization of a high-quality video-stream from the Exposure Rendering framework can be minimized. Exposure Render uses a Monte-Carlo based rendering system to render volumetric data. The earliest estimation of lighting show high degrees of noise, leading to grainy images before...

Numerical Modelling of a Metal Vapour Flow inside a Vapour Distribution Box

Vapour Distribution Boxes (VDBs) are used for continuous Physical Vapour Deposition (PVD) coating. The present study shows a Computational Fluid Dynamics (CFD) simulation of Zinc vapour inside a VDB, analyses the lateral Zinc vapour distribution and identifies regions in which phase change is likely to happen.

An Extensible Toolkit For Real-Time High-Performance Wideband Spectrum Sensing

This document describes the design process of a software toolkit to perform high-performance wideband spectrum sensing. A prominent application of this is Cognitive Radio, a technique that aims to make more efficient use of the available radio spectrum. An extensive theoretical analysis will be performed. Various non-uniform sampling techniques...

De maatschappelijke acceptatie van compressed air energy storage

Bachelorthesis waar de maatschappelijke acceptatie van de energie opslagtechniek compressed air energy storage (CAES) is onderzocht met behulp van Structural Equation Modeling. CAES kan als opslagtechniek bijdragen aan het verhelpen van de variabiliteit en onvoorspelbaarheid van een duurzame energievoorziening. Bij CAES wordt elektriciteit in de...

The behaviour of artificial archaeological remains during a one-dimensional compression test

The Dutch subsoil contains a lot of archaeological remains that need to be preserved for future generations. Until now, it is not possible to predict if these remains will stay intact under the pressure of a manmade structure like an embankment. The goal of this research is to investigate the behaviour of assemblies of artificial archaeological...

1-D Loading of charred particles

The countries who have signed the Malta convention in 1992 are committed to preserve in situ archeological sites instead of excavating them and wasting them for future generations. Unfortunately there are no tools available to predict the impact of civil engineering constructions on buried archeological sites. This report is focused on the...

Dantzig Selector: Het schatten van een sparse parametervector met weinig metingen - Dantzig Selector: Statistical estimation of a sparse vector with undersampled data

Bij lineaire regressie geldt in het algemeen, dat het aantal metingen (n) groter is dan het aantal parameters (p). In praktische situaties kan het echter voorkomen, dat het aantal metingen veel kleiner is dan het aantal parameters. De kleinste kwadratenschatter kan in dat geval niet gebruikt worden om de parametervector te schatten, er is geen...