Biography
I am a fourth-year PhD student in CSE department at UCSD. My advisor is Prof. Ravi Ramamoorthi. Earlier in 2017, I completed my undergraduate in Yao Class, Tsinghua University.
My research interest includes computer graphics and computational photography, with a particular focus on portrait appearance modeling and manipulation.
I am fortunately supported by a Jacobs School of Engineering Fellowship and a Google PhD Fellowship.
News
- [2020.11] Our paper "Neural Light Transport for Relighting and View Synthesis" got accepted by ACM Transactions on Graphics 2020.
- [2020.10] We released our paper "Photon-Driven Neural Path Guiding".
- [2020.9] Google released the feature "Portrait Light" on Pixel 5, which is partly based on our research paper.
- [2020.8] Our paper "Light Stage Super-Resolution: Continuous High-Frequency Relighting" got accepted by SIGGRAPH Asia 2020.
- [2020.6] I started my 15-week internship in Facebook Sausalito remotely with Carlos Aliaga and Christophe Hery.
Publications
Neural Light Transport for Relighting and View Synthesis
Xiuming Zhang, Sean Fanello, Yun-Ta Tsai, Tiancheng Sun, Tianfan Xue, Rohit Pandey, Sergio Orts-Escolano, Philip Davidson, Christoph Rhemann, Paul Debevec, Jonathan T. Barron, Ravi Ramamoorthi, William T. Freeman
ACM Transactions on Graphics 2020
We captured full human bodies with multiple lights and cameras on a light stage, and used a convolutional neural network to predict the texture atlas of known geometric properties, which enables relighting and view synthesis.
Photon-Driven Neural Path Guiding
Shilin Zhu, Zexiang Xu, Tiancheng Sun, Alexandr Kuznetsov, Mark Meyer, Henrik Wann Jensen, Hao Su, Ravi Ramamoorthi
Arxiv
We emit photons from the light sources in the scene, and use a pre-trained neural network to reconstruct the local light fields in the scene, which are then used for path guiding.
Light stage super-resolution: continuous high-frequency relighting
Tiancheng Sun, Zexiang Xu, Xiuming Zhang, Sean Fanello, Christoph Rhemann, Paul Debevec, Yun-Ta Tsai, Jonathan T. Barron, Ravi Ramamoorthi
SIGGRAPH Asia 2020
We use a neural network to super-resolve the lights on the light stage, which enables arbitrary point light relighting on human faces under the light stage.
Single Image Portrait Relighting
Tiancheng Sun, Jonathan T. Barron, Yun-Ta Tsai, Zexiang Xu, Xueming Yu, Graham Fyffe, Christoph Rhemann, Jay Busch, Paul Debevec, Ravi Ramamoorthi
SIGGRAPH 2019
Media coverage, Pixel 5 ''Portrait Light''
We present a system for portrait relighting: a neural network that takes as input a single RGB image of a portrait taken with a standard cellphone camera in an unconstrained environment, and from that image produces a relit image of that subject as though it were illuminated according to any provided environment map.
Connecting Measured BRDFs to Analytic BRDFs by Data-Driven Diffuse-Specular Separation
Tiancheng Sun, Henrik Wann Jensen, Ravi Ramamoorthi
SIGGRAPH Asia 2018
We propose a novel framework for connecting measured and analytic BRDFs, by separating a measured BRDF into diffuse and specular components. This enables measured BRDF editing, a compact measured BRDF model, and insights in relating measured and analytic BRDFs. We also design a robust analytic fitting algorithm for two-lobe materials.
Three-dimensional Display via Multi-layer Translucencies
Tiancheng Sun, Huarong Gu
International Symposium on Optoelectronic Technology and Application 2018
We built a display system using light field display and pyramid-like mirror. The displayed allows the observer to see the virtual 3D objects from different directions in the air.
Attribute‐preserving gamut mapping of measured BRDFs
Tiancheng Sun, Ana Serrano, Diego Gutierrez, Belen Masia
28th Eurographics Symposium on Rendering (EGSR 2017)
1st place at the 2018 ACM Student Research Competition (undergraduate category) ACM annoucement, ACM news, UCSD news
1st place at the SIGGRAPH 2017 ACM Student Research Competition (undergraduate category) SIGGRAPH annoucement, SIGGRAPH blog post, UCSD news
We proposed a new BRDF gamut mapping algorithm using two-step optimization. We optimize the luminance with the guide of perceptual attributes in the first step, and optimize the ink coefficients using image comparison in the second step.
Revisiting Cross-channel Information Transfer for Chromatic Aberration Correction
Tiancheng Sun, Yifan (Evan) Peng, Wolfgang Heidrich
2017 IEEE International Conference on Computer Vision (ICCV)
By modelling the similarity between different channels, we propose a new blind image deconvolution algorithm which transfers information from clear channel to blurry ones, and yield better results than state-of-the-art methods in both refractive and diffractive optical systems.
Convolution Neural Networks with Two Pathways for Image Style Recognition
Tiancheng Sun, Yulong Wang, Jian Yang, and Xiaolin Hu
IEEE Transactions on Image Processing (Volume: 26, Issue: 9, Sept. 2017)
We add the ideas used in image style transfer into traditional CNN network for image style recognition, and achieve state-of-the-art results on three benchmark datasets.
Experience
Research intern @ Facebook 2020.6 - 2020.11
Graphics research at Facebook Reality Lab with Christophe Hery and Carlos Aliaga: we developed a hair inverse rendering system that can infer the geometry and the reflectance of hair strands using images from multiple cameras under different lights.
Research intern @ Google 2019.6 - 2019.9
Computational photography research at Google Research with Jonathan Barron and Graham Fyffe: we super-resolved the light resolution on the light stage so that we could relight portraits under high-frequency lightings.
Software engineering intern @ Google 2018.6 - 2018.9
Computational photography research at Google Research with Yun-Ta Tsai and Jonathan Barron: we developed a relighting system that can change the lighting on a portrait under natural illumination.
Research assistant @ Tsinghua University 2017.2 - 2017.6
Computational display research at State Key Laboratory of Precision Measurement Technology and Instruments with Professor Huarong Gu: we built a 3d volumetric display used light field display and pellicle pyramid.
Research intern @ Universidad de Zaragoza 2016.7 - 2016.10
Computer graphic research at Graphics and Imaging Lab with Professor Diego Gutierrez and Belen Masia: we proposed a new framework for material appearance evaluation, and implemented material editing applications with user-friendlier interface and higher editing capability.
Research intern @ King Abdullah University of Science and Technology 2016.2 - 2016.6
Computational imaging research at Visual Computing Center with Professor Wolfgang Heidrich: we worked on a new approach for image deconvolution, and yielded state-of-the-art results in different optical systems.
Research assistant @ Tsinghua University 2015.12 - 2016.2
Neural network research at State Key Laboratory of Intelligent Technology and Systems with Professor Xiaolin Hu: we proposed a new neural network structure and its corresponding training procedure for image style recognition.
Executive council member @ Tsinghua Spark Program 2015.9 - 2017.6
Organize and plan activities such as Spark Talks and Industrial Field Trips in the program.
Technical leader @ Phantouch Technology 2015.9 - 2015.12
Led a developing group for pose detection on virtual-reality devices.
Research intern @ Megvii Inc. 2015.4 - 2015.9
Computational vision research with Haoqiang Fan: we built a 3d model scanner for human faces based on phase-shift technique.