Supasorn Suwajanakorn

13

Papers

121

Total Citations

Papers (13)

Optimizing Diffusion Noise Can Serve As Universal Motion Priors

DiffusionLight: Light Probes for Free by Painting a Chrome Ball

Diffusion Sampling with Momentum for Mitigating Divergence Artifacts

Zero-shot Inexact CAD Model Alignment from a Single Image

LUSD: Localized Update Score Distillation for Text-Guided Image Editing

Repurposing GANs for One-Shot Semantic Part Segmentation

NeX: Real-Time View Synthesis With Neural Basis Expansion

Diffusion Autoencoders: Toward a Meaningful and Decodable Representation

StyleGAN Salon: Multi-View Latent Optimization for Pose-Invariant Hairstyle Transfer

Learning Geometric-Aware Properties in 2D Representation Using Lightweight CAD Models, or Zero Real 3D Pairs

Guided Motion Diffusion for Controllable Human Motion Synthesis

Zero-guidance Segmentation Using Zero Segment Labels

DiFaReli: Diffusion Face Relighting