Honors Theses
Date of Award
Spring 5-7-2026
Document Type
Undergraduate Thesis
Department
Computer and Information Science
First Advisor
David Harrison
Second Advisor
Yixin Chen
Relational Format
Dissertation/Thesis
Abstract
Signal simulation environments require accurate three dimensional representations of physical spaces, yet current methods for generating these representations, including Light Detection and Ranging (LiDAR) scanning, manual 3D modeling, and commercial photogrammetry, are both costly and time intensive. SpaceForge addresses this gap with a prompt guided pipeline that takes an ordinary indoor photograph and a configurable set of simulation relevant object categories as input and produces a voxelized 3D scene compatible with downstream signal simulation workflows. The pipeline proceeds through five major stages: open set object detection and segmentation, object level preprocessing, single image 3D mesh reconstruction, heuristic pose estimation and scene assembly, and voxelization of the complete scene point cloud. Room geometry, floor, walls, and ceiling, is estimated from the assembled furniture layout and merged into the scene before voxelization, producing a regular voxel grid saved in multiple formats for downstream use. The completed pipeline runs end to end from a single command line call with no manual intervention, completing a typical indoor scene in approximately two to four minutes on a capable GPU.
Recommended Citation
Ross, Compton, "SpaceForge: Spatial Reconstruction for Signal Simulations" (2026). Honors Theses. 3544.
https://egrove.olemiss.edu/hon_thesis/3544