Tianfan Xue
|
|
Tianfan Xue |
Biography
Tianfan Xue started as an Vice Chancellor assistant professor at the Department of Information Engineering, the Chinese University of Hong Kong in Novermber of 2022. If you are excited about computational photography, computer vision, computer graphics, and machine learning, please contact for potential full-funded Ph.D., master, postdoc, R.A., and intern positions available.
Tianfan Xue was working in Google Research from 2017 to 2022. Before that, Tianfan Xue received his Ph.D. from MIT Computer Science and Artificial Intelligence Laboratory, under the supervision of William T. Freeman, with thesis committee members Prof. Frédo Durand, Dr. Richard Szeliski, and Dr. Ce Liu. Also, Tianfan Xue received his Master of Philosophy in the Information Engineering Department, The Chinese University of Hong Kong in 2011, under the supervised of Prof. Xiaoou Tang and Prof. Jianzhuang Liu. Tianfan Xue also received the B.E. degree in the Computer Science and Technology Department from Tsinghua University, China, in 2009. Here is Tianfan Xue's Curriculum Vitae [PDF].Research Highlights
- UltraFusion, an ultra high dynamic range HDR, has won the CVPR 2025 best demo honorable mention.
- BilarfNeRF, an NeRF reconstruction with bilateral grid, has won the SIGGRAPH 2024 best paper honorable mention.
- PhocoLens, an photo-consistent diffusion, has won a spotlight presentation in NeurIPS 2024.
Lab members
Hongyu Deng 邓宏宇 Ph.D. candidate (Co. w. He Chen) |
||||
Postdoc and Visiting Scholars
Alumni
- Yaokun Li (李垚坤)
- Yuehao Wang (王跃豪) , currently is a Ph.D. student at U.T. Austin
- Zhiyi Li (李智毅) , currently is a Ph.D. student at MIT.
- Zhuoya Wang (王卓雅), currently is a master student at Tsinghua University.
Selected Publications
2026 |
|
|
|
FlashVSR: Towards Real-Time Diffusion-Based Streaming Video Super-Resolution |
|
|
DA-VAE: Plug-in Latent Compression for Diffusion via Detail Alignment |
|
|
DynamicTree: Interactive Real Tree Animation via Sparse Voxel Spectrum |
InstantRetouch: Efficient and High-Fidelity Instruction-Guided Image Retouching with Bilateral Space |
|
|
|
CubeComposer: Spatio-Temporal Autoregressive 4K 360° Video Generation from Perspective Video |
|
|
Bi-directional Autoregressive Diffusion for Large Complex Motion Interpolation |
|
|
PhotoFramer: Multi-modal Image Composition Instruction |
Scaling Parallel Sequence Models to Vision Foundation Models |
|
|
|
FullPart: Generating each 3D Part at Full Resolution |
|
|
S2R-HDR: A Large-Scale Rendered Dataset for HDR Fusion |
|
|
LoRA-Edit: Controllable First-Frame-Guided Video Editing via Mask-Aware LoRA Fine-Tuning |
Realtime Video Frame Interpolation using One-Step Diffusion Sampling |
|
Reasoning as Representation: Rethinking Visual Reinforcement Learning in Image Quality Assessment |
|
|
|
ToonComposer: Streamlining Cartoon Production with Generative Post-Keyframing |
2025 |
|
|
|
CamCloneMaster: Enabling Reference-based Camera Control for Video Generation |
|
|
4DSloMo: 4D Reconstruction for High Speed Scene with Asynchronous Capture |
|
|
From One to More: Contextual Part Latents for 3D Generation (CoPart) |
|
|
From Gallery to Wrist: Realistic 3D Bracelet Insertion in Videos |
|
|
GLEAM: Learning Generalizable Exploration Policy for Active Mapping in Complex 3D Indoor Scenes |
AdaptiveAE: An Adaptive Exposure Strategy for HDR Capturing in Dynamic Scenes |
|
Multi-identity Human Image Animation with Structural Video Diffusion |
|
Token-Efficient VLM: High-Resolution Image Understanding via Dynamic Region Proposal |
|
Uni-ISP: Unifying the Learning of ISPs from Multiple Cameras |
|
|
|
UltraFusion: Ultra High Dynamic Imaging using Exposure Fusion CVPR Best Demo Honorable Mention |
Teaching Large Language Models to Regress Accurate Image Quality Scores using Score Distribution |
|
|
|
NVComposer: Boosting Generative Novel View Synthesis with Multiple Sparse and Unposed Images |
PolarFree: Polarization-based Reflection-Free Imaging |
|
Unifying Appearance Codes and Bilateral Grids for Driving Scene Gaussian Splatting |
|
|
|
A 3D-Aware and Controllable Framework for Cinematic Text-to-Video Generation |
A Physics-Informed Blur Learning Framework for Imaging Systems |
|
2024 |
|
Event-assisted 12-stop HDR Imaging of Dynamic Scene |
|
|
|
PhoCoLens: Photorealistic and Consistent Reconstruction in Lensless Imaging |
|
|
Goal Conditioned Reinforcement Learning for Photo Finishing Tuning |
AdaptiveISP: Learning an Adaptive Image Signal Processor for Object Detection |
|
HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation |
|
|
|
DualDn: Dual-domain Denoising via Differentiable ISP |
Event-Based Motion Magnification |
|
|
|
Descriptive Image Quality Assessment in the Wild Depicting Beyond Scores: Advancing Image Quality Assessment through Multi-modal Language Models |
TimeLens-XL: Real-time Event-based Video Frame Interpolation with Large Motion |
|
AutoDIR: Automatic All-in-One Image Restoration with Latent Diffusion |
|
|
|
Bilateral Guided Radiance Field Processing SIGGRAPH Honorable Mention |
|
|
LenslessFace : An End-to-End Optimized Lensless System for Privacy-Preserving Face Verification |
GenNBV: Generalizable Next-Best-View Policy for Active 3D Reconstruction |
|
Text-to-3D Generation with Bidirectional Diffusion using both 2D and 3D priors |
|
Interactive3D: Create What You Want by Interactive 3D Generation |
|
HDRFlow: Real-Time HDR Video Reconstruction with Large Motions |
|
EmbodiedScan: A Holistic Multi-Modal 3D Perception Suite Towards Embodied AI |
|
Before 2024 |
|
Obj-NeRF: Extracting Object NeRFs from Multi-view Images |
|
AligNeRF: High-Fidelity Neural Radiance Fields via Alignment-Aware Training |
|
Fast and High-Quality Image Denoising via Malleable Convolutions |
|
How to Train Neural Networks for Flare Removal |
|
Defocus Map Estimation and Deblurring from a Single Dual-Pixel Image |
|
Learned Dual-View Reflection Removal |
|
Joint Bilateral Learning for Real-time Universal Photorealistic Style Transfer Real-time Localized Photorealistic Video Style Transfer. |
|
Handheld Mobile Photography in Very Low Light |
|
Multi-frame stereo matching with edges, planes, and superpixels, |
|
Unprocessing Images for Learned Raw Denoising, |
|
Stereoscopic Dark Flash for Low-light Photography, |
|
MoSculp: Interactive Visualization of Shape and Time, |
|
Video Enhancement with Task-Oriented Flow, |
|
Pix3D: Dataset and Methods for Single-Image 3D Shape Modeling |
|
Best-Buddies Similarity - Robust Template Matching using Mutual Nearest Neighbors, |
|
MarrNet: 3D Shape Reconstruction via 2.5D Sketches |
|
|
Visual Dynamics: Probabilistic Future Frame Synthesis via Cross Convolutional Networks Visual dynamics: Stochastic future generation via layered cross convolutional networks |
|
Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling |
|
Single Image 3D Interpreter Network 3D Interpreter Networks for Viewer-Centered Wireframe Modeling International Journal of Computer Vision (IJCV), 2018. [arxiv-preprint] |
|
A Computational Approach for Obstruction-Free Photography |
|
The Aperture Problem for Refractive Motion |
|
Refraction Wiggles for Measuring Fluid Depth and Velocity from Video |
|
Example-Based 3D Object Reconstruction from Line Drawings
|
|
|
Symmetric Piecewise Planar Object Reconstruction from a Single Image 3D Modeling from a Single View of a Symmetric Object |
|
Object Cut: Complex 3D object reconstruction through line drawing separation |
|
Joint Example-based DepthMap Super-Resolution |
|
Fast Frame-rate Up-conversion of Depth Video via Video Coding |
|
|
* indicates equal contribution
|
Education
Ph.D., Computer Sci., Massachusetts Institute of Technology, Aug. 2012 - Jul. 2017
Supervisor: Prof. William T. Freeman.
M.Phil., Information Eng., Chinese University of Hong Kong, Aug. 2009 - Jul. 2011
Supervisor: Prof. Xiaoou Tang
B. Eng., Computer Sci. & Tech, Tsinghua University, China, Aug. 2005 - Jul. 2009
Services
Conference Organizers: Computer Vision and Pattern Recognition 2020 web chair Computer Vision and Pattern Recognition 2023, 2024 area chair Winter Conference on Applications of Computer Vision 2023 area chair
Conference Reviewers: Computer Vision and Pattern Recognition (CVPR): 2016, 2017, 2018, 2020, 2021, 2022 European Conference on Computer Vision (ECCV): 2016, 2018, 2020, 2022, workshop on optical flow (2018) Pacific Graphics 2018 International Conference on Computer Vision (ICCV) 2017, 2019, 2021 SIGGRAPH 2018, 2021, 2022 SIGGRAPH Asia 2017, 2021, 2022 Neural Information Processing Systems (NeurIPS) 2016, 2021, 2022 International Symposium on Circuits & Systems (ISCAS) 2017
Journal Reviewers: IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI) IEEE Transactions on Systems, Man, and Cybernetics IEEE Transactions on Computational Imaging (TCI) IEEE Transactions on Multimedia (TMM) IEEE Transactions on Image Processing (TIP) IEEE Transactions on Circuits and Systems for Video Technology (TCSVT) Artificial Intelligence Image and Vision Computing (IVC) Cognitive Computation Computers and Electrical Engineering Machine Vision and Applications IEEE Computer Graphics and Applications Journal of the Optical Society of America Pattern Recognition Letter Optics and Lasers in Engineering
MISC
Previous Labmates in MIT CSAIL: Katie Bouman, Hossein Mobahi, Michael Rubinstein, Tali Dekel, Donglai Wei, Jiajun Wu, Xiuming Zhang, Zhoutong Zhang