Vilma Ylvén and Oscar Montelin present their Master's thesis Using NeRF- and Mesh-Based Methods to Improve Visualisation of Point Clouds on Friday 12 January at 10:15 in MH:309A

Abstract engelska:
In recent years the field of synthetic view points has seen some major improvements. Most importantly with the publication of Neural Radiance Fields allowing for hugely detailed and accurate 3D novel views. Usage of LiDAR products to collect actual depth data has also seen an increase as it is immensely useful for achieving high resolution 3D mapping of a space. However, these point clouds can be hard to read as they give a discrete sample of surfaces and lack colour and texture. In this thesis we have explored various ways of improve visualisation and human understanding of scenes and objects generated from a stationary camera/LiDAR pair. We have done this by firstly isolating individual rigid moving object in a scene and constructing denser point clouds of these objects by projecting them on the video and aggregating over time. By utilising the novel view synthesis method Point-NeRF [14] we have also tried to improve visualisation of these dense point clouds further. This has been done by training a point-based neural network on the aggregated point clouds and the corresponding video frames. Lastly two methods for surface reconstruction of objects and the backgrounds have been tested. With this we have achieved accurate and understandable renders of a variety of vehicles. We believe that with a well calibrated camera this method shows significant promise for reconstructing scenes in 3D in post-processing well.

Supervisors:
Kalle Åström, supervisor, Centre for Mathematical Sciences
Johanna Engman, co-supervisor, Centre for Mathematical Sciences

Examiner:
Magnus Oskarsson, examiner, Centre for Mathematical Sciences