Pieter Peers

Pieter Peers is an Associate Professor in the Department of Computer Science at the College of William & Mary. Prior to this, he was a senior researcher at the Graphics Laboratory at the University of Southern California's Institute for Creative Technologies and a Research Assistant Professor at USC's Viterbi School of Engineering.

His educational background includes a Ph.D. in Computer Science from K.U.Leuven (Belgium) in 2006 and a BS/MS in the same field from the same institution in 2000.

Peers' research focuses on relighting, appearance modeling, and neural rendering—central topics in computer graphics and vision. His work bridges traditional rendering techniques with modern deep learning, particularly in inverse rendering, material modeling, and image-based relighting. He has made significant contributions to subsurface scattering, photometric stereo, and neural rendering systems.

His recent publications, including work in SIGGRAPH, CVPR, and ICCV, demonstrate a strong trend toward integrating diffusion models and transformer architectures into appearance modeling and relighting tasks. These works often address real-world challenges such as single-image relighting, material estimation, and lighting control with minimal user input.

• RenderFormer (SIGGRAPH 2025): Transformer-based neural rendering with global illumination.
• ScribbleLight (CVPR 2025): Scribble-driven indoor relighting using diffusion models.
• MatFusion (SIGGRAPH Asia 2023): Diffusion-based SVBRDF capture.

He has mentored and collaborated with numerous students and researchers, including Chong Zeng, Jun Myeong Choi, and James Bieron, indicating active involvement in advising and research supervision. While no formal grants or awards are listed, his consistent publication record in top venues suggests sustained research funding and recognition in the graphics community.

Peers has led research in advanced rendering systems, including work on compressive light transport sensing, wavelet environment matting, and compact representations of heterogeneous subsurface scattering. His early work at USC and K.U.Leuven laid the foundation for his current research in neural appearance modeling and inverse rendering.