Scholar
Jonathan Granskog
Google Scholar ID: LxrjN68AAAAJ
Runway
Citations & Impact
Citations
800
H-index
6
i10-index
5
Publications
9
Co-authors
0
Publications
Resume
Research Experience
- Worked on a new implementation for Runway's video-to-video product based on the latest Gen-3 Alpha model.
- Collaborated with intern Shailesh Mishra to extend 'neural neighbor feature matching' to work with CLIP-ResNet50’s feature maps and allow using multiple style images during optimization.
- Participated in Hendrik Baatz's Master's thesis project, which was later continued during an internship at NVIDIA Research, presenting a method of applying 'mesoscale' materials using neural fields onto 3D objects.
- Started a project on neural scene graph rendering near the end of his internship at NVIDIA, which proposed a novel neural scene representation inspired by traditional scene graphs.
- Explored how learned neural scene representations could be used to augment traditional graphics pipelines during his Master's thesis, which was further developed during an internship at NVIDIA Zurich.
Background
- Research interests include video-to-video conversion, 3D asset stylization, and the application of neural fields to 3D objects. Specializes in computer graphics and deep learning.
Miscellany
- Personal interests may include advancements in computer graphics rendering technologies and innovation in related fields.