IEEE

ChildDiffusion: Unlocking the Potential of Generative AI and Controllable Augmentations for Child Facial Data Using Stable Diffusion and Large Language Models

Abstract: Ensuring the availability of child facial datasets is essential for advancing AI applications, yet legal, ethical, and data scarcity concerns pose significant challenges. Current generative ...
GitHub

DAPS: Improving Diffusion Inverse Problem Solving with Decoupled Noise Annealing (CVPR 2025 Oral)

We propose a new method called Decoupled Annealing Posterior Sampling (DAPS) that relies on a novel noise annealing process to solve posterior sampling with diffusion prior. Specifically, we decouple ...
IEEE

TSPDiffuser: Diffusion Models as Learned Samplers for Traveling Salesperson Path Planning Problems

Abstract: This paper presents TSPDiffuser, a novel data-driven path planner for traveling salesperson path planning problems (TSPPPs) in environments rich with obstacles. Given a set of destinations ...
GitHub

Adaptive Non-Uniform Timestep Sampling for Diffusion Model Training

This repository contains the official implementation of (CVPR 2025) Adaptive Non-Uniform Timestep Sampling for Diffusion Model Training by Myunsoo Kim*, Donghyeon Ki*, Seong-Woong Shim, and Byung-Jun ...