Zhanghao Sun

I am a CV/ML researcher at Apple, working in image/video/3D compression and reconstruction.

I obtained my Ph.D. degree from Electrical Engineering, Stanford University, where I worked in the field of computational imaging hardware and algorithms. I was advised by Prof. Olav Solgaard, and co-advised by Prof. Gordon Wetzstein.

In Summer 2019, I worked as an algorithm engineer intern at Adaps, a startup that manufactures direct time-of-flight (dToF) sensors. In Summer 2021, I worked as a research intern at Snapchat, computational imaging team. In Summer 2022, I worked as a research intern at Meta Reality Labs, on-device computer vision team. In Summer 2023, I worked as a research intern at NextCam team at Adobe, led by Professor Marc Levoy.

Email / CV / Google Scholar / Linkedin / Github

Selected Publications

	High Contrast Nulling in Photonic Meshes Through Architectural Redundancy Carson Valdez, Zhanghao Sun, Anne R Kroo, David AB Miller, Olav Solgaard Optics Letters, 2025 paper We demonstrated world-record 100dB extinction ratio in photonics imaging system.


	AstroPIC: near-infrared photonic integrated circuit coronagraph architecture for the Habitable Worlds Observatory Dan Sirbu, Ruslan Belikov, Kevin Fogarty, Carson Valdez, Zhanghao Sun, Annie Kroo, Olav Solgaard, David AB Miller, Olivier Guyon Space Telescopes and Instrumentation, 2024 paper We proposed extremely high-contrast astronomical photonics imaging system for exo-planet observatory (in collaboration with NASA!).


Click to view difference z	Structured Light with Redundancy Codes Zhanghao Sun, Xinxin Zuo, Dong Huo, Yu Zhang, Yiming Qian, Jian Wang Optics Express, 2024 paper code We adopt redundancy coding from the communication system to improve the robustness of the structured light system.


	Scalable low-latency optical phase sensor array Zhanghao Sun, Sunil Pai, Carson Valdez, Maziyar Milanizadeh, Andrea Melloni, Francesco Morichetti, David A. B. Miller, and Olav Solgaard Optica, 2023 paper We designed and demonstrated reference-free scalable phase sensor array based on photonics circuit.


	Consistent Direct Time-of-Flight Video Depth Super-Resolution Zhanghao Sun, Wei Ye, Jinhui Xiong, Gyeongmin "Chris" Choe, Jialiang Wang, Shuochen Su, Rakesh Ranjan CVPR, 2023 paper supplementary project code video We propose algorithm and dataset for high accuracy, temporally consistent time-of-flight video depth super-resolution.
	Energy-Efficient Adaptive 3D Sensing Brevin Tilmon, Zhanghao Sun, Sanjeev Koppal, Yicheng Wu, Georgios Evangelidis, Ramzi Zahreddine, Guru Krishnan, Sizhuo Ma, Jian Wang CVPR, 2023 project code We propose energy-efficient and eye-safe active 3D sensor with parallel adaptive sampling.
Click to view difference	Seeing Far in the Dark with Patterned Flash Zhanghao Sun, Jian Wang, Yicheng Wu, Shree Nayar ECCV, 2022 paper supplementary poster code video Our proposed novel flash pattern shows significantly better imaging quality at long distances in low-light environments.


	Transparent Camera Jian Wang, Zhanghao Sun, Dejia Xu ICCP, 2022 (Best Poster Award)* poster We proposed the transparent camera prototype with potential application in mobile devices.
	Experimentally realized in situ backpropagation for deep learning in nanophotonic neural networks Sunil Pai, Zhanghao Sun, Tyler W Hughes, Taewon Park, Ben Bartlett, Ian Williamson, Momchil Minkov, Maziyar Milanizadeh, Nathnael Abebe, Francesco Morichetti, Andrea Melloni, Shanhui Fan, Olav Solgaard, David Miller. Science, 2023 paper code For the first time, we realized neural network back-propagation in the optics (photonics) domain.
	Resonant scanning design and control for fast spatial sampling Zhanghao Sun, Ronal Quan, Olav Solgaard Scientific Reports, 2021 CLEO (Oral), 2021 journal conference code Fast, adaptive spatial sampling with resonant scanner for better machine perception and localization.




	SPADnet: deep RGB-SPAD sensor fusion assisted by monocular depth estimation Zhanghao Sun, David Lindell, Olav Solgaard, Gordon Wetzstein Optics Express, 2020 paper project code We propose a neural network solution to fuse monocular depth and metric depth from noisy dToF sensors.