Songcen Xu

10

Papers

90

Total Citations

Papers (10)

Co-Speech Gesture Video Generation via Motion-Decoupled Diffusion Model

MirrorGaussian: Reflecting 3D Gaussians for Reconstructing Mirror Reflections

MagicEraser: Erasing Any Objects via Semantics-Aware Control

Self-Adaptive Reality-Guided Diffusion for Artifact-Free Super-Resolution

TextField3D: Towards Enhancing Open-Vocabulary 3D Generation with Noisy Text Fields

Semantics-aware Motion Retargeting with Vision-Language Models

VastGaussian: Vast 3D Gaussians for Large Scene Reconstruction

BIVDiff: A Training-Free Framework for General-Purpose Video Synthesis via Bridging Image and Video Diffusion Models

DreamControl: Control-Based Text-to-3D Generation with 3D Self-Prior

Any-Size-Diffusion: Toward Efficient Text-Driven Synthesis for Any-Size HD Images