I am a research associate in Prof. Dieter Schmalstieg's group at the VISUS Lab of the University of Stuttgart, specializing in controllable neural shading for real-time rendering. I received my Ph.D. from the State Key Lab of CAD&CG at Zhejiang University, where I was advised by Prof. Rui Wang and Prof. Yuchi Huo. During my doctoral studies, I was fortunate to collaborate with Prof. Evan Peng at the University of Hong Kong and Dr. Yazhen Yuan at Tencent.
Warning
Problem: The current name of your GitHub Pages repository ("Solution: Please consider renaming the repository to "
http://".
However, if the current repository name is intended, you can ignore this message by removing "{% include widgets/debug_repo_name.html %}" in index.html.
Action required
Problem: The current root path of this site is "baseurl ("_config.yml.
Solution: Please set the
baseurl in _config.yml to "Research Interests
My doctoral work focuses on high-quality frame generation techniques, using data-driven methods to accelerate real-time rendering of 3D scenes with complex motion and diverse lighting conditions in interactive applications.
To date, I have worked on spatiotemporal supersampling for real-time rendering, neural light transport, and generative models for image synthesis.
If you are interested in potential research collaborations, please feel free to contact me.
For beginners If you are just beginning your research journey or going through a difficult period, I would be very happy to share my experiences and offer any support I can. Please do not hesitate to reach out.News
Publications ( * equal contribution, # corresponding author)
Selected
DISK: Differentiable Sparse Kernel Complex for Efficient Spatially-Variant Convolution
Zhizhen Wu*, Zhe Cao*, Yuchi Huo#
ICLR 2026
A differentiable framework that decomposes arbitrary dense kernels into sparse layers and interpolates them for spatially varying filtering, using 1/100 of the optimization iterations of the prior SOTA and achieving up to 20x speedup on mobile.
DISK: Differentiable Sparse Kernel Complex for Efficient Spatially-Variant Convolution
Zhizhen Wu*, Zhe Cao*, Yuchi Huo#
ICLR 2026
Consecutive Frame Extrapolation with Predictive Sparse Shading
Zhizhen Wu, Zhe Cao, Yazhen Yuan, Zhilong Yuan, Rui Wang, Yuchi Huo#
ACM Transactions on Graphics (SIGGRAPH Asia 2025)
A predictive sparse shading framework that jointly learns extrapolation error, flows, and frame synthesis to selectively shade dynamic regions while reusing generated pixels elsewhere.
Consecutive Frame Extrapolation with Predictive Sparse Shading
Zhizhen Wu, Zhe Cao, Yazhen Yuan, Zhilong Yuan, Rui Wang, Yuchi Huo#
ACM Transactions on Graphics (SIGGRAPH Asia 2025)
MoFlow: Motion-Guided Flows for Recurrent Rendered Frame Prediction
Zhizhen Wu, Zhilong Yuan, Chenyu Zuo, Yazhen Yuan, Yifan Peng, Guiyang Pu, Rui Wang, Yuchi Huo#
ACM Transactions on Graphics (Vol.44, No.22, 2025)
A temporally coherent motion representation and estimator that constructs motion-guided flows from recurrent latent features, enabling stable multi-frame extrapolation.
MoFlow: Motion-Guided Flows for Recurrent Rendered Frame Prediction
Zhizhen Wu, Zhilong Yuan, Chenyu Zuo, Yazhen Yuan, Yifan Peng, Guiyang Pu, Rui Wang, Yuchi Huo#
ACM Transactions on Graphics (Vol.44, No.22, 2025)
Adaptive Recurrent Frame Prediction with Learnable Motion Vectors
Zhizhen Wu, Chenyu Zuo, Yuchi Huo#, Yazhen Yuan, Yifan Peng, Guiyang Pu, Rui Wang, Hujun Bao
SIGGRAPH Asia 2023 (Conference Paper)
A rendered frame extrapolation framework that learns residual flow over rendered motion vectors for low-latency recurrent prediction, handling occlusions, dynamic shading, and translucent objects.
Adaptive Recurrent Frame Prediction with Learnable Motion Vectors
Zhizhen Wu, Chenyu Zuo, Yuchi Huo#, Yazhen Yuan, Yifan Peng, Guiyang Pu, Rui Wang, Hujun Bao
SIGGRAPH Asia 2023 (Conference Paper)
All Publications
DPF: Differentiable Polyphase Filtering for Large-Kernel Approximation
Zhe Cao*, Zhizhen Wu*, Zhonggui Chen, Rui Wang, Yuchi Huo#
SIGGRAPH 2026 (Conference Paper)
A differentiable multi-resolution polyphase filter using spatial-to-depth transformation, intensity-decoupled sampling, and layer-adaptive filtering, with 2x single-kernel and 2.5x spatially varying speedups over the prior SOTA.
DPF: Differentiable Polyphase Filtering for Large-Kernel Approximation
Zhe Cao*, Zhizhen Wu*, Zhonggui Chen, Rui Wang, Yuchi Huo#
SIGGRAPH 2026 (Conference Paper)
StereoFG: Generating Stereo Frames from Centered Feature Stream
Chenyu Zuo, Yazhen Yuan, Zhizhen Wu, Zhijian Liu, Jingzhen Lan, Ming Fu, Yuchi Huo, Rui Wang
SIGGRAPH Asia 2025 (Conference Paper)
A stereo frame generation pipeline for VR that aligns binocular features into centered stream to upsample alternating low-resolution views into temporally stable high-quality stereo frames.
StereoFG: Generating Stereo Frames from Centered Feature Stream
Chenyu Zuo, Yazhen Yuan, Zhizhen Wu, Zhijian Liu, Jingzhen Lan, Ming Fu, Yuchi Huo, Rui Wang
SIGGRAPH Asia 2025 (Conference Paper)
IntrinsicControlNet: Cross-distribution Image Generation with Real and Unreal
Jiayuan Lu, Rengan Xie, Zixuan Xie, Zhizhen Wu, Dianbing Xi, Qi Ye, Rui Wang, Hujun Bao, Yuchi Huo
ICCV 2025
A diffusion framework that takes intrinsic images (geometry, materials, and lighting) as controls to disentangle synthetic control signals from real-image content for controllable photorealistic image generation.
IntrinsicControlNet: Cross-distribution Image Generation with Real and Unreal
Jiayuan Lu, Rengan Xie, Zixuan Xie, Zhizhen Wu, Dianbing Xi, Qi Ye, Rui Wang, Hujun Bao, Yuchi Huo
ICCV 2025
NeLT: Object-oriented Neural Light Transfer
Chuankun Zheng*, Yuchi Huo*, Shaohua Mo, Zhihua Zhong, Zhizhen Wu, Wei Hua, Rui Wang, Hujun Bao#
ACM Transactions on Graphics (Vol.42, No.162, 2023)
An object-oriented neural light transfer representation that decomposes scene global illumination into per-object neural transport functions, enabling interactive global illumination for dynamic lighting, materials, and geometry without path-traced inputs.
NeLT: Object-oriented Neural Light Transfer
Chuankun Zheng*, Yuchi Huo*, Shaohua Mo, Zhihua Zhong, Zhizhen Wu, Wei Hua, Rui Wang, Hujun Bao#
ACM Transactions on Graphics (Vol.42, No.162, 2023)