Fangzhou Hong

19

Papers

122

Total Citations

Papers (19)

LN3Diff: Scalable Latent Neural Fields Diffusion for Speedy 3D Generation

Generative Gaussian Splatting for Unbounded 3D City Generation

EgoLM: Multi-Modal Language Model of Egocentric Motions

GeneMAN: Generalizable Single-Image 3D Human Reconstruction from Multi-Source Human Data

CityDreamer: Compositional Generative Model of Unbounded 3D Cities

Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation

LiDAR-Based Panoptic Segmentation via Dynamic Shifting Network

Versatile Multi-Modal Pre-Training for Human-Centric Perception

DeformToon3D: Deformable Neural Radiance Fields for 3D Toonification

ReMoDiffuse: Retrieval-Augmented Motion Diffusion Model

SHERF: Generalizable Human NeRF from a Single Image

HuMMan: Multi-modal 4D Human Dataset for Versatile Sensing and Modeling

3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion

EgoLife: Towards Egocentric Life Assistant

MEAT: Multiview Diffusion Model for Human Generation on Megapixels with Mesh Attention

SurMo: Surface-based 4D Motion Modeling for Dynamic Human Rendering

Garment4D: Garment Reconstruction from Point Cloud Sequences

PrimDiffusion: Volumetric Primitives Diffusion for 3D Human Generation

4D Panoptic Scene Graph Generation