Description
Difficult surfaces like glass, mirrors, and marble make laser scans inaccurate or even impossible. We will analyze laser scans in waveform mode to see if we can find detectable patterns in the signal. If there are, we will classify points as outliers to improve scan quality.
Tasks
Depending on the type of thesis/project, some of these:
- Data acquisition with our laser scanner
- Waveform pattern analysis
- Dataset processing and cleaning
- Waveform classification via machine learning
Requirements
- Knowledge of English language (source code comments and final report should be in English)
- Knowledge of Python, Numpy, Scipy, Pytorch
- Knowledge of deep-learning basics
Environment
The project should be implemented platform independent (Linux, Windows).
A bonus of €500/€1000 if completed to satisfaction within an agreed time-frame of 6/12 months (PR/BA or DA)
Additional Images and Files
Attachment | Size |
---|---|
Photo with mirror | 2.83 MB |
Reconstruction with duplicated geometry | 1.66 MB |
Responsible
For more information please contact Philipp Erler.