Yang Li 교수

We present Lepard, a Learning based approach for partial point cloud matching in rigid and deformable scenes. The key characteristics are the following techniques that exploit 3D positional knowledge for point cloud matching: 1) An architecture that disentangles point cloud representation into feature space and 3D position space. 2) A position encoding method that explicitly reveals 3D relative distance information through the dot product of vectors. 3) A repositioning technique that modifies th

Tracking non-rigidly deforming scenes using range sensors has numerous applications including computer vision, AR/VR, and robotics. However, due to occlusions and physical limitations of range sensors, existing methods only handle the visible surface, thus causing discontinuities and in-completeness in the motion field. To this end, we introduce 4DComplete, a novel data-driven approach that estimates the non-rigid motion for the unobserved geometry. 4DComplete takes as input a partial shape and

Understanding the fundamental processes of human brain development and diseases is of great importance for our health. However, existing research models such as non-human primate and mouse models remain limited due to their developmental discrepancies compared with humans. Over the past years, an emerging model, the "brain organoid" integrated from human pluripotent stem cells, has been developed to mimic developmental processes of the human brain and disease-associated phenotypes to some extent

Maximizing the potential of human liver organoids (LOs) for modeling human septic liver requires the integration of innate immune cells, particularly resident macrophage Kupffer cells. In this study, we present a strategy to generate LOs containing Kupffer cells (KuLOs) by recapitulating fetal liver hematopoiesis using human induced pluripotent stem cell (hiPSC)-derived erythro-myeloid progenitors (EMPs), the origin of tissue-resident macrophages, and hiPSC-derived LOs. Remarkably, LOs actively

One of the widespread solutions for non-rigid tracking has a nested-loop structure: with Gauss-Newton to minimize a tracking objective in the outer loop, and Preconditioned Conjugate Gradient (PCG) to solve a sparse linear system in the inner loop. In this paper, we employ learnable optimizations to improve tracking robustness and speed up solver convergence. First, we upgrade the tracking objective by integrating an alignment data term on deep features which are learned end-to-end through CNN.

Liver disease is a global health issue that has caused an economic burden worldwide. Organ transplantation is the only effective therapy for end-stage liver disease; however, it has been hampered by a shortage of donors. Human pluripotent stem cells (hPSCs) have been widely used for studying liver biology and pathology as well as facilitating the development of alternative therapies. hPSCs can differentiate into multiple types of cells, which enables the generation of various models that can be

We present a computational approach that implements the time-dependent complete-active-space self-consistent-field method, as introduced in [Phys. Rev. A 88, 023402 (2013)]. Our implementation addresses the challenge of diatomic molecules subjected to an intense laser pulse by considering the full dimensionality of the problem using prolate spheroidal coordinates. The method incorporates the gauge-invariant frozen-core approximation, boosts the evaluation of the electron-electron interaction ter

Mutations of the OPA1 gene are responsible for over 70% of autosomal dominant optic atrophy patients. Peripheral blood mononuclear cells (PBMCs) were isolated from a 27-year-old patient with heterozygous c.2708_2711delTTAG mutation in the OPA1 gene. PBMCs were reprogrammed into induced pluripotent stem cell (iPSC) line with episomal plasmids encoding hOCT4, hSOX2, hNANOG, hLIN28, hKLF4 and hL-MYC. The established iPSC line had normal karyotype, expressed pluripotent markers, and was capable to d

We have successfully implemented the time-dependent complete-active-space self-consistent-field method on prolate spheroidal coordinates. This will open a way to first-principle study of strong-field and attosecond phenomena in diatomic molecules.

We present an implementation of a time-dependent multiconfiguration self-consistent-field method [R. Anzaki, T. Sato, and K. L. Ishikawa, Phys. Chem. Chem. Phys. 19, 22008 (2017)] with the full configuration-interaction expansion for coupled electron-nuclear dynamics in diatomic molecules subject to a strong laser field. In this method, the total wave function is expressed as a superposition of different configurations constructed from time-dependent electronic Slater determinants and time-depen

Integration of resident immune cells into in vitro organoid models is important for accurately recapitulating native tissue physiology. Here, we present a protocol for integrating liver-resident macrophages, Kupffer cells, into liver organoid models derived from human induced pluripotent stem cells (iPSCs). We describe procedures for generating Kupffer cell progenitors and hepatic endoderm from iPSCs, followed by detailed steps for establishing liver organoids containing Kupffer cells (KuLOs). T