Zheng's Horadric Cube
- about-me
- a researcher working at NVIDIA Applied Deep Learning Research
- previously worked with
- Prof. Lingqi Yan at UCSB
- Marco Salvi, Lifan Wu, Jan Novák, Daqi Lin, Prof. Ravi Ramamoorthi, and Miloš Hašan at NVIDIA Real-time Graphics Research
- Daqi Lin, Markus Kettunen, Chris Wyman, Lifan Wu, and Prof. Ravi Ramamoorthi at NVIDIA Real-time Graphics Research
- Miloš Hašan, Valentin Deschaintre, Iliyan Georgiev, Yannick Hold-Geoffroy, Yiwei Hu, and Fujun Luan at Adobe Research
- Edward Liu at NVIDIA Applied Deep Learning Research
- Professors Lu Wang, Yanning Xu, and Beibei Wang at SDU
- research goal: ultimate photorealism at minimal cost
- regular-reading
- principles of effective research - Michael A. Nielsen
- the craft of writing effectively - Larry McEnerney
- the importance of deconstruction - Kilian Q. Weinberger
- See more ...
- publications
-
rgbx-next: towards realistic generative rendering from g-buffers
Zheng Zeng, Marco Salvi, Lifan Wu, Jan Novák, Daqi Lin, Saeed Hadadan, Yichen Sheng, Robert Pottorff, Shiqiu Liu, Ravi Ramamoorthi, Lingqi Yan, Miloš HašanarXiv preprint, 2026 -
videomatgen: pbr materials through joint generative modeling
-
videoneumat: neural material extraction from generative video models
Bowen Xue, Saeed Hadadan, Zheng Zeng, Fabrice Rousselle, Zahra Montazeri, Miloš HašanACM SIGGRAPH 2026 (Conference track) -
restir pg: path guiding with spatiotemporally resampled paths
"restir gives you ideal guiding distributions for free (?)"ACM SIGGRAPH Asia 2025 (Conference Track) -
diffusion-guided relighting for single-image svbrdf estimation
ACM SIGGRAPH Asia 2025 (Conference Track) -
reshadable level of detail impostors for fast distant objects rendering
Eurographics Symposium on Rendering 2025 (CGF Track) -
rgb↔x: image decomposition and synthesis using material- and lighting-aware diffusion models
"diffusion models can eliminate/simulate gi"Zheng Zeng, Valentin Deschaintre, Iliyan Georgiev, Yannick Hold-Geoffroy, Yiwei Hu, Fujun Luan, Lingqi Yan, Miloš HašanACM SIGGRAPH 2024 (Conference Track) -
extrass: a framework for joint spatial super sampling and frame extrapolation
"you can jointly upsample & extrapolate frames"Songyin Wu, Sungye Kim, Zheng Zeng, Deepak Vembar, Sangeeta Jha, Anton Kaplanyan, Lingqi YanACM SIGGRAPH Asia 2023 (Conference Track) -
ray-aligned occupancy map array for fast approximate ray tracing
-
lightweight neural basis functions for all-frequency shading
"neural basis functions act as real ones, but faster"ACM SIGGRAPH Asia 2022 -
temporally reliable motion vectors for better use of temporal information
-
temporally reliable motion vectors for real-time ray tracing
"different motion vectors for different effects"Eurographics 2021 (CGF track) -
joint svbrdf recovery and synthesis from a single image using an unsupervised generative adversarial network
"no corresponding svbrdfs? still can train a gan!"Eurographics Symposium on Rendering 2020 (DL-only Track) -
denoising stochastic progressive photon mapping renderings using a multi-residual network
"different receptive fields for different scales of noise"Journal of Computer Science and Technology (CVM2020)
