I am a senior engineer at Qualcomm AI Research in San Diego, working on 3D computer vision and machine learning. I received my PhD in EE from the University of Notre Dame under Dr. Nicholas Zabaras on deep learning for modeling PDE systems. I studied automatic control at Tongji University.

I'm interested in 3D reconstruction, inverse rendering, generative design, and XR applications. I also worked on data compression.


OpenShape teaser

Scaling Up 3D Shape Representation Towards Open-World Understanding
Minghua Liu*, Ruoxi Shi*, Kaiming Kuang*, Yinhao Zhu, Xuanlin Li, Shizhong Han, Hong Cai, Fatih Porikli, Hao Su

Scale up object-level point cloud pretraining (global embedding) with point-text-image contrastive learning on Objaverse. Data enigeering and scaling matter.

FIPT teaser

FIPT: Factorized Inverse Path Tracing for Efficient and Accurate Material-Lighting Estimation
Liwen Wu*, Rui Zhu*, Mustafa B. Yaldiz, Yinhao Zhu, Hong Cai, Janarbek Matai, Fatih Porikli, Tzu-Mao Li, Manmohan Chandraker, Ravi Ramamoorthi
ICCV 2023 Project

Resolve the ambiguity of material and lighting estimation with precomputed radiance transfer (factorized), differentiable path tracing, radiance caching and material segmentation.

PartSLIP instance segmentation

PartSLIP: Low-Shot Part Segmentation for 3D Point Clouds via Pretrained Image-Language Models
Minghua Liu, Yinhao Zhu, Hong Cai, Shizhong Han, Zhan Ling, Fatih Porikli, Hao Su
CVPR 2023 Project

Lift the amazing low-shot 2D part detection capability of GLIP to 3D point cloud, together with point oversegmentation and multiview fusion.

rate-distortion-computation tradeoff
Transformer-based Transform Coding
Yinhao Zhu*, Yang Yang*, Taco Cohen
ICLR 2022 OpenReview

AFAIK SwinT-ChARM is the first neural image codec that outperforms VTM in rate-distortion while with comparable decoding time on GPU.

PLONQ bitstream

Progressive Neural Image Compression with Nested Quantization and Latent Ordering
Yadong Lu*, Yinhao Zhu*, Yang Yang*, Amir Said, Taco Cohen
ICIP 2021 arXiv  |  IEEE

An embedded bitstream is obtained with nested quantization and per-element sorting by prior stddev, based on the hyperprior model.

PReLU transforms for coding YUV

Transform network architectures for deep learning based end-to-end image/video coding in subsampled color spaces
Hilmi Egilmez, Ankitesh Singh, Muhammed Coban, Marta Karczewicz, Yinhao Zhu, Yang Yang, Amir Said, Taco Cohen
IEEE Open Journal of Signal Processing, arXiv

PReLU can replace GDN in the hyperprior model to compress YUV (and RGB!) images without loss of coding gain.

cGlow prediction
Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data
Yinhao Zhu, Nicholas Zabaras, Phaedon-Stelios Koutsourelakis, Paris Perdikaris
JCP 2019  |  arXiv  |  Code
A Poisson-Gaussian Denoising Dataset with Real Fluorescence Microscopy Images
Yide Zhang*, Yinhao Zhu*, Evan Nichols, Qingfei Wang, Siyuan Zhang, Cody Smith, Scott Howard
CVPR 2019  |  arXiv  |  Code
Deep convolutional encoder-decoder networks for uncertainty quantification of dynamic multiphase flow in heterogeneous media
Shaoxing Mo, Yinhao Zhu, Nicholas Zabaras, Xiaoqing Shi, Jichun Wu
WRR 2018  |  arXiv  |  Code
Bayesian Deep Convolutional Encoder-Decoder Networks for Surrogate Modeling and Uncertainty Quantification
Yinhao Zhu, Nicholas Zabaras
JCP 2018  |  arXiv  |  Code
Performance guaranteed human-robot collaboration through correct-by-design
Xiaobin Zhang, Yinhao Zhu, Hai Lin
ACC 2016


ZBOT: a two-wheeled self-balancing robot
Undergrad project, 2012