AI model learns subsurface scattering for realistic object relighting

Researchers have developed a novel method to capture and represent subsurface scattering properties of light transport with high detail. This technique utilizes a U-Net Convolutional Neural Network (CNN) trained on data from a stereo projector-camera setup employing phase-shifted profilometry patterns. The model reconstructs dense pixel footprint responses, enabling realistic relighting of objects with arbitrary high-resolution projector patterns, and has demonstrated generalization to unseen materials. AI

IMPACT This research could lead to more realistic rendering and virtual object manipulation in computer graphics and augmented reality applications.

RANK_REASON The cluster contains an academic paper detailing a new method for computer vision research. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• phase-shifted profilometry (PSP) patterns
• pixel footprint response
• stereo projector-camera setup
• subsurface scattering

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →