Wenzheng Chen

The Less You Depend, The More You Learn: Synthesizing Novel Views from Sparse, Unposed Images without Any 3D Knowledge

FieryGS: In-the-Wild Fire Synthesis with Physics-Integrated Gaussian Splatting

Value-State Gated Attention for Mitigating Extreme-Token Phenomena in Transformers

Parametric SDF for Dynamic Surface Reconstruction

Blind Inversion using Latent Diffusion Priors

GeoSplating: Towards Geometry Guided Gaussian Splatling for Physically-based Inverse Rendering