(English) Results

Results from the project include:

• Neural Rendering for Synthetic Scenes:
  • Active Exploration for Neural Global Illumination of Variable Scenes: please see project page here.
  • Neural Precomputed Radiance Transfer: publication page  here
• Neural Rendering & Relighting for Captured Scenes & Faces:
  • Point-Based Neural Rendering with Per-View Optimization: Please see project page here
  • FreeStyleGAN: Free-view Editable Portrait Rendering with the Camera Manifold: Please see project page here.
  • Free-viewpoint Indoor Neural Relighting from Multi-view Stereo: Please see project page here
  • Multi-view relighting using geometry and deep learning: Please see the project page here.
• Image- and Video-Based Rendering and Editing:
  • Video-Based Rendering of Dynamic Stationary Environments from Unsynchronized Inputs: please see publication here
  • Image-Based Rendering of Cars using Semantic Labelling: Please see publication here.
  • Realistic Compositing of Image-Based Scenes: Please see publication here.

• Material Capture and Transfer:
  • Multi-image SVBRDF recovery: Please see publication here.
  • Guided fine tuning for large scale material transfer: Please see publication here.
• Global Illumination:
  • Product Path guiding: Please see publication here.
• Contrast enhancement for VR: Please see publication here (and project page at Cambridge).