GPU-enabled lego brick approximation of mesh models

• Main Authors: Canilao, Ivan Ray T., Pamplona, Patrick Karlo N., Vallesteros, Daniel Martin H.
• Format: text
• Language: English
• Published: Animo Repository 2011
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/11844
• Institution: De La Salle University

Bautista (2007) developed an application that approximates a Lego model based on an input 3D mesh model. The Lego model which is made up of Lego bricks is essentially an approximation of the mesh model. The quality of the approximation depends on the resolution. A higher resolution results into a more detailed Lego model. However, as the resolution increases, the executive time also increased. This research is concerned with GPU enabled Lego brick approximation of mesh models. The research proponents implemented the approximation algorithm using GPU + CPU. GPU programming using NVIDIAs CUDA allows faster execution than a CPU only resolution. The research proponents discovered that when the GPUs multi-processing power was not utilized enough, it was not always the case that GPU-CPU is faster than a CPU implementation. Though the larger part of the data collected shows that the GPU+CPU implementation achieves up to 4x speedup compared to the CPU only implementation if there were intensive processing to be done.