Po-Nan Li

About me

With the training in signal processing and optics, Po-Nan works at the intersection of imaging and machine learning. Having extensive internship experience in the VR/AR industry, Po-Nan now works on smartphone cameras, aiming to make the already amazing cameras in your pocket even better.

Experience

Camera Machine Learning Engineer

Apple 2021-Present

Machine learning solutions for camera hardware design and image quality analysis.

Research Intern

Facebook Reality Labs 2020

Machine learning frameworks for accelerated nanostructure waveguide design.

Hardware Engineering Intern

Google 2019

Algorithm development and camera calibration instrumentation for VR/AR technologies.

Publications

Sequence-guided protein structure determination using graph convolutional and recurrent networks

Po-Nan Li, Saulo de Oliveira, Soichi Wakatsuki, and Henry van den Bedem

The 20th IEEE International Conference on BioInformatics and BioEngineering (2020)

Hybrid real- and reciprocal-space full-field imaging with coherent illumination

Po-Nan Li, Soichi Wakatsuki, Piero Pianetta, and Yijin Liu

J. Opt. 22, 115611 (2020)

A bacterial surface layer protein exploits multi-step crystallization for rapid self-assembly

Jonathan Herrmann, Po-Nan Li, Fatemeh Jabbarpour, Anson C. K. Chan, Ivan Rajkovic, Tsutomu Matsui, Lucy Shapiro, John Smit, Thomas M. Weiss, Michael E. P. Murphy, and Soichi Wakatsuki

Proc. Natl. Acad. Sci. U.S.A. 117, 388 (2019)

Transport Properties of Nanoporous, Chemically Forced Biological Lattices

Po-Nan Li, Jonathan Herrmann, Soichi Wakatsuki, and Henry van den Bedem

J. Phys. Chem. B 123, 10331 (2019)

Holographic Phase Retrieval and Reference Design

D.A. Barmherzig, J. Sun, P.-N. Li, and E.J. Candès

Inverse Problems 35, 9 (2019)

Nutrient transport suggests an evolutionary basis for charged archaeal surface layer proteins

Po-Nan Li, Jonathan Herrmann, Bradley B. Tolar, Frédéric Poitevin, Rasika Ramdasi, John R. Bargar, David A. Stahl, Grant J. Jensen, Christopher A. Francis, Soichi Wakatsuki, and Henry van den Bedem

ISME J. 12, 2389 (2018)

Determination of three-dimensional atomic positions from tomographic reconstruction using ensemble empirical mode decomposition

Po-Nan Li, Zong-Han Wu, Chien-Nan Hsiao, Ting-Kuo Lee, and Chien-Chun Chen

New J. Phys. 18, 083025 (2016)

Method to enhance the resolution of x-ray coherent diffraction imaging for non-crystalline bio-samples

Ti-Yen Lan, Po-Nan Li, and Ting-Kuo Lee

New J. Phys. 16, 033016 (2014)

Subwavelength localization of near fields in coupled metallic spheres for single-emitter polarization analysis

Po-Nan Li, Hsiu-Hao Tsao, Jer-Sing Huang, and Chen-Bin Huang

Opt. Lett. 36, 2339-2341 (2011)

Projects

Taiwanese Referendum 2018 - An interactive website that visualizes the results of 2018 Referendum in Taiwan. Contributed to data crawling and front-end Development.

CS231n: Peak finding for crystallography - Built and trained a YOLO network that can rapidly locate Bragg peaks on a diffraction image on SLAC's massive SFX data.

EE367: Mountable Dynamic Range Enhancer for Digital Cameras - Built an Arduino-enabled device that can be mounted onto a camera and enhance the photo dynamic range by adaptively modulating a gray-scale LCD.

Leadership

Regional Director

North American Taiwan Studies Association Annual Conference 2017

Coordinated meal, accomadation and venue logistics for a 4 day conference with more than 100 international attendees.

Vice President

Stanford Taiwanese Student Association (STSA) 2016

Co-led the 100+ member organization and held events for the biggest Taiwanese community in the Silicon Valley.