paper-with-me

Papers

NoiseCtrl: A Sampling-Algorithm-Agnostic Conditional Generation Method for Diffusion Models

2025-01-01 · CVPR 2025 1 · Longquan Dai, He Wang, Jinhui Tang

In training-free conditional generative tasks, diffusion models utilize differentiable loss functions to steer the generative reverse process, necessitating modifications to sampling algorithms like DDPM and DDIM. However, such adjustments likely reduce flexibility and reliability. In this paper, we propose NoiseCtrl, a sampling-algorithm-agnostic technique for controlled image generation. Essentially, diffusion models generate denoised results z_ t-1 by adding a predicted mean \mu_ t with random noise epsilon_ t . NoiseCtrl specifically adjusts the random noise while leaving the underlying sampling algorithms unchanged. At each step t , NoiseCtrl converts the unconditional Gaussian noise into conditional noise epsilon'_ t by substituting the isotropic Gaussian distribution with the von Mises-Fisher distribution. This substitution introduces a directional focus while preserving the randomness required for conditional image generation. Thanks to this non-intrusive design, NoiseCtrl is straightforward to integrate and has been extensively validated through experiments, demonstrating its adaptability for different diffusion algorithms and superior performance across various conditional generation tasks.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Conditional Image GenerationImage Generation

Methods 이 논문이 사용한 방법론

Focus 설명 없음
Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

MolMiner: Towards Controllable, 3D-Aware, Fragment-Based Molecular Design

2024-11-10 · Raul Ortega-Ochoa, Tejs Vegge, Jes Frellsen

We introduce MolMiner, a fragment-based, geometry-aware, and order-agnostic autoregressive model for molecular design. MolMiner supports conditional generation of molecules over twelve properties, enabling flexible contr…

3D geometryBenchmarking

Token Perturbation Guidance for Diffusion Models

2025-06-10 · Javad Rajabi, Soroush Mehraban, Seyedmorteza Sadat, Babak Taati

Classifier-free guidance (CFG) has become an essential component of modern diffusion models to enhance both generation quality and alignment with input conditions. However, CFG requires specific training procedures and i…

Ctrl-Z Sampling: Diffusion Sampling with Controlled Random Zigzag Explorations

2025-06-25 · Shunqi Mao, Wei Guo, Chaoyi Zhang, Weidong Cai

Diffusion models have shown strong performance in conditional generation by progressively denoising Gaussian noise toward a target data distribution. This denoising process can be interpreted as a form of hill climbing i…

Denoising

Energy-based models for diagnostic reconstruction and analysis in a laboratory plasma device

2026-05-09 · Phil Travis, Troy Carter arxiv

Energy-based models (EBMs) provide a powerful and flexible way of learning a joint probability distribution over data by constructing an energy surface. This energy surface enables insight extraction and conditional samp…

Anomaly Detection

Segmentation-Guided Knee Radiograph Generation using Conditional Diffusion Models

2024-04-04 · Siyuan Mei, Fuxin Fan, Fabian Wagner, Mareike Thies 외

Deep learning-based medical image processing algorithms require representative data during development. In particular, surgical data might be difficult to obtain, and high-quality public datasets are limited. To overcome…

Segmentation