Föllmer processes are variationally optimal among generative diffusions because they minimize the impact of drift estimation error on path-space KL divergence, rendering different interpolation schedules statistically equivalent.
hub
i2- sb: Image-to-image schr \” odinger bridge
18 Pith papers cite this work. Polarity classification is still indexing.
18
Pith papers citing it
hub tools
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Empirical study finds synthetic-to-real domain gap sharply degrades diffusion SR models on real cross-sensor satellite pairs while real-data training faces optimization and adaptation problems.
Timage generates text query overlays on images via Constrained Schrödinger Bridge to boost fine-grained spatial reasoning in vision-language models, outperforming larger systems on VMCBench with a 7B backbone.
SKILD unifies unconditional image generation and continuous super-resolution in one diffusion model via scale-invariant k-space dynamics where the reverse process handles both tasks by varying only the starting timestep.
DBMSolver is a new training-free sampler using exponential integrators that reduces NFEs by up to 5x and improves quality in diffusion bridge model-based image-to-image translation tasks.
FMA introduces flow matching for multi-step cross-modal feature alignment in few-shot learning, using fixed coupling, noise augmentation, and early-stopping to outperform one-step PEFT methods.
Stochastic interpolants unify flow-based and diffusion-based generative models by bridging target densities exactly via latent-variable processes whose drifts minimize quadratic objectives.
WFM achieves near-diffusion quality for all four BraTS MRI modalities with one 82M model in 1-2 steps by flowing from the mean of conditioning modalities in wavelet space, running 250-1000x faster.
LiveMoments restores reselected key photos in Live Photos via reference-guided diffusion and motion alignment, yielding higher perceptual quality and fidelity than prior methods especially under fast motion.
MaskDiME uses adaptive masked diffusion to produce 30x faster, localized, and semantically consistent visual counterfactual explanations without training, matching or exceeding prior performance on five datasets.
DPDL learns multiple Gaussian prototypes and a Schrödinger bridge diffusion process to enclose normal samples in a compact discriminative space while using hyperspherical dispersion to identify out-of-distribution anomalies, reporting SOTA results on 9 datasets.
AnF-DiffPET is a CT-conditioned diffusion framework for low-dose PET denoising incorporating anatomical-frequency guidance, multi-scale cross-transformer reconstruction, and frequency-contrastive hard mining, showing gains over prior CNN, GAN, transformer, and diffusion methods on four datasets.
VDSB-GWSyn uses DSB conditioned on vessel masks and a shape prior to synthesize guidewires, yielding downstream localization gains when used for pre-training.
FluxFlow uses conservative pixel-space flow-matching with uncertainty weights and Wiener test-time correction to outperform baselines on photometric and scientific accuracy for ground-to-space super-resolution, validated on a new real 19,500-pair DESI-HST dataset.
A conditional diffusion model trained on partitioned incomplete samples for physical dynamics achieves asymptotic convergence to the true generative process under mild conditions and outperforms baselines in imputation.
A survey that introduces taxonomies for categorizing pre-trained diffusion model methods applied to inverse problems and analyzes their connections and challenges.
NaviEdit is a training-free inference-time controller that decouples edit progress from model scale traversal in diffusion-based image editing via self-consistency, reporting average gains across editors and backbones.
citing papers explorer
-
Variational Optimality of F\"ollmer Processes in Generative Diffusions
Föllmer processes are variationally optimal among generative diffusions because they minimize the impact of drift estimation error on path-space KL divergence, rendering different interpolation schedules statistically equivalent.
-
Mind the Gap: Quantifying the Domain Gap in Cross-Sensor Diffusion Super-Resolution
Empirical study finds synthetic-to-real domain gap sharply degrades diffusion SR models on real cross-sensor satellite pairs while real-data training faces optimization and adaptation problems.
-
Timage: A Generative Text-in-Image Paradigm for Fine-Tuning Vision-Language Models
Timage generates text query overlays on images via Constrained Schrödinger Bridge to boost fine-grained spatial reasoning in vision-language models, outperforming larger systems on VMCBench with a 7B backbone.
-
Everything at Every Scale: Scale-Invariant Diffusion with Continuous Super-Resolution
SKILD unifies unconditional image generation and continuous super-resolution in one diffusion model via scale-invariant k-space dynamics where the reverse process handles both tasks by varying only the starting timestep.
-
DBMSolver: A Training-free Diffusion Bridge Sampler for High-Quality Image-to-Image Translation
DBMSolver is a new training-free sampler using exponential integrators that reduces NFEs by up to 5x and improves quality in diffusion bridge model-based image-to-image translation tasks.
-
Exploring Cross-Modal Flows for Few-Shot Learning
FMA introduces flow matching for multi-step cross-modal feature alignment in few-shot learning, using fixed coupling, noise augmentation, and early-stopping to outperform one-step PEFT methods.
-
Stochastic Interpolants: A Unifying Framework for Flows and Diffusions
Stochastic interpolants unify flow-based and diffusion-based generative models by bridging target densities exactly via latent-variable processes whose drifts minimize quadratic objectives.
-
WFM: 3D Wavelet Flow Matching for Ultrafast Multi-Modal MRI Synthesis
WFM achieves near-diffusion quality for all four BraTS MRI modalities with one 82M model in 1-2 steps by flowing from the mean of conditioning modalities in wavelet space, running 250-1000x faster.
-
LiveMoments: Reselected Key Photo Restoration in Live Photos via Reference-guided Diffusion
LiveMoments restores reselected key photos in Live Photos via reference-guided diffusion and motion alignment, yielding higher perceptual quality and fidelity than prior methods especially under fast motion.
-
MaskDiME: Adaptive Masked Diffusion for Precise and Efficient Visual Counterfactual Explanations
MaskDiME uses adaptive masked diffusion to produce 30x faster, localized, and semantically consistent visual counterfactual explanations without training, matching or exceeding prior performance on five datasets.
-
Distribution Prototype Diffusion Learning for Open-set Supervised Anomaly Detection
DPDL learns multiple Gaussian prototypes and a Schrödinger bridge diffusion process to enclose normal samples in a compact discriminative space while using hyperspherical dispersion to identify out-of-distribution anomalies, reporting SOTA results on 9 datasets.
-
AnF-DiffPET: Anatomy- and Frequency-Guided Diffusion for PET/CT Denoising
AnF-DiffPET is a CT-conditioned diffusion framework for low-dose PET denoising incorporating anatomical-frequency guidance, multi-scale cross-transformer reconstruction, and frequency-contrastive hard mining, showing gains over prior CNN, GAN, transformer, and diffusion methods on four datasets.
-
VDSB-GWSyn: Diffusion Schr\"{o}dinger Bridge for Controllable and Anatomically Feasible Guidewire Synthesis in Coronary Angiography
VDSB-GWSyn uses DSB conditioned on vessel masks and a shape prior to synthesize guidewires, yielding downstream localization gains when used for pre-training.
-
FluxFlow: Conservative Flow-Matching for Astronomical Image Super-Resolution
FluxFlow uses conservative pixel-space flow-matching with uncertainty weights and Wiener test-time correction to outperform baselines on photometric and scientific accuracy for ground-to-space super-resolution, validated on a new real 19,500-pair DESI-HST dataset.
-
Incomplete Data, Complete Dynamics: A Diffusion Approach
A conditional diffusion model trained on partitioned incomplete samples for physical dynamics achieves asymptotic convergence to the true generative process under mild conditions and outperforms baselines in imputation.
-
A Survey on Diffusion Models for Inverse Problems
A survey that introduces taxonomies for categorizing pre-trained diffusion model methods applied to inverse problems and analyzes their connections and challenges.
-
Semantic Granularity Navigation in Image Editing
NaviEdit is a training-free inference-time controller that decouples edit progress from model scale traversal in diffusion-based image editing via self-consistency, reporting average gains across editors and backbones.
- Unifying Deep Stochastic Processes for Image Enhancement