one-step-diffusion-model topic

Repo for SeedVR2 & SeedVR (CVPR2025 Highlight)

ByteDance-Seed

SeedVR2: One-Step Video Restoration via Diffusion Adversarial Post-Training

IceClear

(CVPR 2025) Adversarial Diffusion Compression for Real-World Image Super-Resolution [PyTorch]

Guaishou74851

Author's Implementation for E-LatentLPIPS

mingukkang

Any-to-Bokeh is a novel one-step video bokeh framework that converts arbitrary input videos into temporally coherent, depth-aware bokeh effects.

vivoCameraResearch

[CVPR 2025] MoFlow: One-Step Flow Matching for Human Trajectory Forecasting via Implicit Maximum Likelihood Estimation Distillation

DSL-Lab