Hi, I am a 4th year PhD Candidate at Beihang University, my supervisor is Prof. Jianhui Zhao.
I also work closely with Prof. Yebin Liu at Tsinghua University since 2014.
My research area is Computer Vision and Computer Graphics.
Specifically, I'm working on Real-time Dynamic 3D Reconstruciton Algorithms for Human Performance Capture.
We propose DeepHuman, a deep learning based framework for 3D human reconstruction from a single RGB image. Since this problem is highly intractable, we adopt a stage-wise, coarse-to-fine method consisting of three steps, namely inner body estimation, outer surface reconstruction and frontal surface detail refinement.
IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2019)
This paper proposes a new method for live free-viewpoint human performance capture with dynamic details (e.g., cloth wrinkles) using a single RGBD camera. Our main contributions are: (i) a multi-layer representation of garments and body, and (ii) a physics-based performance capture procedure.
European Conference on Computer Vision (ECCV 2018)
We propose a light-weight and highly robust real-time human performance capture method based on hybrid depth&IMU sensors. The method can reconstruct challenging motions, detailed geometries and the inner human body shapes of a clothed subject simultaneously in real-time.
IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2018 Oral Presentation)
We propose DoubleFusion, a new real-time system that combines volumetric dynamic reconstruction with datadriven template ﬁtting to simultaneously reconstruct detailed geometry, non-rigid motion and the inner human body shape from a single depth camera.
IEEE International Conference on Computer Vision (ICCV 2017)
We propose BodyFusion, a novel real-time geometry fusion method that can track and reconstruct non-rigid surface motion of a human performance using a single consumer-grade depth camera.
ACM Transactions on Graphics (Present in SIGGRAPH 2017)
This paper proposes a real-time method that uses a single-view RGBD input to simultaneously reconstruct a casual scene with a detailed geometry model, surface albedo, per-frame non-rigid motion and per-frame low-frequency lighting, without requiring any template or motion priors.