J
Jinli Suo
Researcher at Tsinghua University
Publications - 152
Citations - 3494
Jinli Suo is an academic researcher from Tsinghua University. The author has contributed to research in topics: Computer science & Pixel. The author has an hindex of 28, co-authored 124 publications receiving 2587 citations. Previous affiliations of Jinli Suo include Chinese Academy of Sciences & MediaTech Institute.
Papers
More filters
Proceedings ArticleDOI
Efficient 3D kernel estimation for non-uniform camera shake removal using perpendicular camera system
Tao Yue,Jinli Suo,Qionghai Dai +2 more
TL;DR: An acceleration method to compute the 3D projection of 2D local blur kernels fast, and then derive the3D kernel by interpolating from a minimal set of local blurernels, based on the minimal 3D kernel solver.
Proceedings ArticleDOI
Computational Hyperspectral Imaging
TL;DR: Computational imaging as discussed by the authors is a field of optical systems in which an image is not formed by a lens and simply sampled onto the detector, but the process of image formation is facilitated by both the power of the optical elements and the computational processing of the sampled light signal.
Proceedings ArticleDOI
Recent advances of deep learning for spectral snapshot compressive imaging
TL;DR: In this paper , a review of deep learning methods for spectral snapshot compressive imaging (SCI) is presented, which used a single shot measurement to capture the three-dimensional (3D, x, y, λ) spectral image.
Proceedings ArticleDOI
Learning-based ray sampling strategy for computation efficient neural radiance field generation
Yuqi Han,Jinli Suo,Qionghao Dai +2 more
TL;DR: In this paper , a learning-based sampling strategy is proposed to conduct dense sampling in regions with rich texture and sparse sampling in other regions, extremely reducing the computation resources and accelerating the learning speed.
HOPE: High-order Polynomial Expansion of Black-box Neural Networks
TL;DR: High-order Polynomial Expansion (HOPE) as mentioned in this paper is a method for expanding a neural network into a high-order Taylor polynomial on a reference input, which provides an explicit expression of the network's local interpretations.