Animatable Implicit Neural Representations for Creating Realistic Avatars From Videos.

This paper addresses the challenge of reconstructing an animatable human model from a multi-view video. Some recent works have proposed to decompose a non-rigidly deforming scene into a canonical neural radiance field and a set of deformation fields that map observation-space points to the canonical space, thereby enabling them to learn the dynamic scene from images. However, they represent the deformation field as translational vector field or SE(3) field, which makes the optimization highly under-constrained. Moreover, these representations cannot be explicitly controlled by input motions. Instead, we introduce blend weight fields to produce the deformation fields. Based on the skeleton-driven deformation, blend weight fields are used with 3D human skeletons to generate observation-to-canonical and canonical-to-observation correspondences. Since 3D human skeletons are more observable, they can regularize the learning of deformation fields. Moreover, the blend weight fields can be combined with input skeletal motions to generate new deformation fields to animate the human model. To improve the quality of human modeling, we further represent the human geometry as a signed distance field in the canonical space. Additionally, a neural point displacement field is introduced to enhance the capability of the blend weight field on modeling detailed human motions. Experiments show that our approach significantly outperforms recent human modeling methods.

Implicit Neural Representations With Structured Latent Codes for Human Body Modeling.

This paper addresses the challenge of novel view synthesis for a human performer from a very sparse set of camera views. Some recent works have shown that learning implicit neural representations of 3D scenes achieves remarkable view synthesis quality given dense input views. However, the representation learning will be ill-posed if the views are highly sparse. To solve this ill-posed problem, our key idea is to integrate observations over video frames. To this end, we propose Neural Body, a new human body representation which assumes that the learned neural representations at different frames share the same set of latent codes anchored to a deformable mesh, so that the observations across frames can be naturally integrated. The deformable mesh also provides geometric guidance for the network to learn 3D representations more efficiently. Furthermore, we combine Neural Body with implicit surface models to improve the learned geometry. To evaluate our approach, we perform experiments on both synthetic and real-world data, which show that our approach outperforms prior works by a large margin on novel view synthesis and 3D reconstruction. We also demonstrate the capability of our approach to reconstruct a moving person from a monocular video on the People-Snapshot dataset.

Shape Prior Guided Instance Disparity Estimation for 3D Object Detection.

In this paper, we propose a novel system named Disp R-CNN for 3D object detection from stereo images. Many recent works solve this problem by first recovering point clouds with disparity estimation and then apply a 3D detector. The disparity map is computed for the entire image, which is costly and fails to leverage category-specific prior. In contrast, we design an instance disparity estimation network (iDispNet) that predicts disparity only for pixels on objects of interest and learns a category-specific shape prior for more accurate disparity estimation. To address the challenge from scarcity of disparity annotation in training, we propose to use a statistical shape model to generate dense disparity pseudo-ground-truth without the need of LiDAR point clouds, which makes our system more widely applicable. Experiments on the KITTI dataset show that, when LiDAR ground-truth is not used at training time, Disp R-CNN outperforms previous state-of-the-art methods based on stereo input by 20 percent in terms of average precision for all categories. The code and pseudo-ground-truth data are available at the project page: https://github.com/zju3dv/disprcnn.

Shear stress improves the endothelial progenitor cell function via the CXCR7/ERK pathway axis in the coronary artery disease cases.

BACKGROUND: Dysfunction in the late Endothelial Progenitor Cells (EPCs) is responsible for endothelial repair in patients with Coronary Artery Disease (CAD), and the shear stress is beneficial for EPCs function. However, the impact of shear stress on the capacity of EPCs in CAD patients has not been elucidated yet. The C-X-C chemokine receptor 7/extracellular signal-regulated kinase (CXCR7)/(ERK) pathways are identified to regulate EPCs function in CAD patients. Here, we hypothesize that shear stress upregulates the CXCR7/ERK pathways, which restore the EPCs function in CAD patients. METHODS: The human Peripheral Blood Mononuclear Cells (PBMCs) were collected from healthy adults and CAD patients and then used for EPCs cultivation. The Lv-siRNA for human CXCR7 was transfected into induced EPCs isolated from the CAD patients. Meanwhile, the EPCs from CAD patients were subjected to shear stress generated by a biomimetic device. Next, the cell viability, migration, tube formation, and apoptosis were detected by CCK-8, Transwell assay, Matrigel, and flow cytometry, respectively. Also, the CXCR7/ERK pathways in human EPCs were analyzed by Western blotting and qRT-PCR. RESULT: Compared to the EPCs collected from normal adults, the CAD patient-derived EPCs showed reduced in vitro vasculogenic capacity. Also, the level of CXCR7 in CAD patient-derived EPCs was significantly reduced compared to the EPCs of healthy subjects. Meanwhile, the extracellular signal-regulated kinase (ERK), which represents a CXCR7 downstream signaling pathway, had decreased phosphorylation level. The shear stress treatment augmented the CXCR7 expression and also elevated ERK phosphorylation, which is comparable to the up-regulation of CAD patient-derived EPCs function. Further, the small interfering RNA (siRNA)-mediated CXCR7 knockdown diminished the enhanced migration, adhesion, and tube formation capacity of shear stress treated CAD patient-derived EPCs. CONCLUSION: Up-regulation of the CXCR7/ERK pathways by shear stress can be a promising new target in enhancing the vasculogenic ability of CAD patient-derived EPCs.

High-fat-diet-induced obesity upregulates the expression of lymphoid chemokines and promotes the formation of gastric lymphoid follicles after Helicobacter suis infection.

Helicobacter suis colonizes the stomachs of a variety of animals, including humans, and is more likely than other Helicobacter species to induce gastric mucosa-associated lymphoid tissue lymphoma. Obesity is a low-grade chronic inflammatory state in which the induction of a chemokine network contributes to a variety of diseases. However, the effect of obesity on the development of gastric MALT in the presence of H. suis infection remains unclear. Here, we reveal that high-fat-diet-induced obesity upregulates the expression of lymphoid chemokines in the stomach and accelerates the H. suis infection-induced formation of gastric lymphoid follicles, potentially via a mechanism that involves the activation of the nuclear factor kappa B (NF-κB) signaling pathway. These findings provide novel insight into the pathogenesis of obesity-related diseases, especially those induced by Helicobacter infection.

Saccharomyces boulardii administration can inhibit the formation of gastric lymphoid follicles induced by Helicobacter suis infection.

Helicobacter suis has a greater tendency to induce gastric mucosa-associated lymphoid tissue lymphoma compared with other Helicobacter species in humans and animals. Saccharomyces boulardii has been established as an adjunct to H. pylori eradication treatment, but the effect of S. boulardii administration alone on Helicobacter infection remains unclear. Here, we found that S. boulardii administration effectively decreased the bacterial load of H. suis and inhibited the formation of lymphoid follicles in the stomach post-infection. The levels of H. suis-specific immunoglobulin A (IgA) and secretory IgA in the gastric juice and small intestinal secretions and the production of mouse ß-defensin-3 in the small intestinal secretions were significantly increased by S. boulardii administration at 12 weeks after H. suis infection. In addition, feeding with S. boulardii inhibited the expression of inflammatory cytokines and lymphoid follicle formation-related factors after H. suis infection. These results suggested that S. boulardii may be useful for the prevention and treatment of Helicobacter infection-related diseases in humans.