Papers

Field

Top tools here

Open problems in Diffusion priors

Balancing initialization and stochasticity in diffusion-based inverse problems1 paper raises it · 1 addresses it
Excessive noise during sampling can erase gains from good initialization, degrading reconstruction quality.
- FLINT: Coupling Proximal Initialization and Bounded Stochasticity for Flow-Matching Inverse Problemsraises · addresses
Joint image-measurement consistency in inverse problems1 paper raises it · 1 addresses it
Most methods condition on noisy measurements but do not enforce consistency between reconstructed image and denoised measurement.
- Dual Flow: Joint Image-Measurement Flow for Inverse Problemsraises · addresses
Handling sparse, asynchronous, or missing sensor data in real-time reconstruction1 paper raises it · 1 addresses it
Real-world sensing systems often have irregular or incomplete measurements, limiting applicability of standard methods.
- Fault-tolerant shallow recurrent decoders for sparse sensing with missing sensor measurementsraises · addresses
Task-driven measurement design for single-pixel imaging1 paper raises it · 1 addresses it
Fixed illumination patterns may not be optimal for classification tasks, wasting acquisition time and photons.
- LIPS-PC: Learning Illumination Patterns for Single-Pixel Classificationraises · addresses
Generalization sensitivity to injected noise in diffusion sampling1 paper raises it · 1 addresses it
Small changes in noise during sampling can break strong generalization across models, affecting reproducibility.
- The role of injected noise in diffusion model generalizationraises · addresses

Named by an AI from these papers' abstracts.

Open problems

Recently funded

Topics in this field

computed tomography · ct reconstruction · computed tomography cbct79 papers17 this week
diffusion priors · diffusion prior · imaging inverse77 papers12 this week

Topics in this field

Active contributors · 90 days

Active contributors

Activity · 12 weeks

Activity

Recent papers

All papers in Research

arXiv cs.CVOct 31
3D Neural Rendering for Automated Ultrasonic Laser Scans
Thomas Pierson, Matthew Mitchell, Kira Perna, Anand Iyer, Eric Flynn +1
Laser doppler vibrometer (LDV) scans can be used to detect defects and cracks in manufactured parts before they are deployed. This work presents a method for reconstructing the steady-state surface wavefield of a three-dimensional part by combining automated robotic manipulation with neural rendering. A six-degree-of-freedom (6-DOF) robotic manipulator (UFactory xArm6) with an embedded piezoelectric transducer presented an acoustically excited test article to a stationary LDV at a sequence of known orientations, yielding a set of complex-valued 2D scans with poses derived analytically from the arm kinematics rather than from image-based pose estimation. These scans were used to train a novel complex-nerfacto model, a modified Neural Radiance Field (NeRF) pipeline that replaces the standard three-channel RGB output head with a two-channel output. Complex-nerfacto represents the real and imaginary velocity components of a wavefield, preserving the complex measurement without quantization. Reconstructions of a welded aluminum cylinder and a welded T-bracket were evaluated against measured scans using Z-normalized cross-correlation (ZNCC). The complex-valued pipeline achieved a mean…
PDF arXiv
arXiv cs.CVOct 30
Learning Clean Generative Processes from Noisy Measurements
Fabian Kruse, Lorenzo Baldassari, Ivan Dokmanić
We study how to learn a generative model of clean images using only noisy observations. Diffusion processes and their generalizations to non-Gaussian noise require posterior predictions at every noise level. We first show that it is possible to learn these predictions from independent noisy views using Noise2Noise regression, but only at or above the noise level of the training data. To learn the predictions below that level, needed to generate clean images, we impose consistency: a prediction should equal its expectation after continuing the sampling process. We derive conditions under which consistency uniquely determines the continuation to the clean image distribution and verify them for Poisson, binomial, Gamma, and negative-binomial measurements. Experiments on CIFAR-10, photon-counting microscopy, and synthetic aperture radar show that the required posterior predictions can be learned from noisy views alone and that the consistency relation repairs the continuation beyond the noise level of the data, which severely deteriorates without consistency.
PDF arXiv
arXiv eess.IVOct 30
Does Vendor Image Processing Matter? Learning Echocardiography Display Appearance from Native Beamspace Data
Elias Stenhede, Arian Ranjbar
Most echocardiography datasets contain vendor-processed images, but processing is often undocumented and may affect model performance and robustness. To address this, we train a small convolutional neural network to reproduce General Electric display appearance from native beamspace brightness-mode ultrasound, using 8 406 recordings from 605 unique patients. The beamspace CNN reduces test mean absolute error by 43% over the optimal pointwise brightness map, outperforms an equivalent Cartesian CNN and runs in real time on a CPU. Across the EchoXFlow and CAMUS datasets, unfiltered training followed by filtered evaluation yields the lowest mean Dice for every class, suggesting filtering may be useful as augmentation when evaluation data are filtered or their processing is unknown. Code and pretrained weights: https://github.com/Ahus-AIM/EchoAppearance
PDF arXiv
arXiv stat.MLOct 30
The Prior-Force Family of Implicit Denoiser Priors and Its Decisions in Linear Inverse Problems
Weijie Liang, Chuling Wen, Jian Lu
Implicit priors given by Gaussian denoisers successfully regularize a wide range of inverse problems in approaches such as plug-and-play (PnP), regularization by denoising (RED) and score-based methods. Such a prior depends on the noise level, and how it selects among signals consistent with the data remains largely uncharacterized. To put implicit and explicit priors on the same footing, we define the prior-force family $W_t=t^{-1}(D_t^{-1}-I)$ of any Gaussian denoiser $D_t$ at noise variance $t>0$ (inverse in the graph sense), the normalized denoising residual as an operator on clean signals. At a common effective prior weight, the equilibrium equations differ only in their prior terms: PnP uses $W_t$, RED/score-based methods the score force (the normalized residual as an operator on noisy inputs), and maximum a posteriori (MAP) estimation the MAP force (the negative log-prior gradient). For the minimum mean-square error (MMSE) denoisers of a prior $π$, we compare the corresponding energies at three levels. Without observation, raising the noise level favors PnP minima that retain more posterior variance. On signals consistent with the data, the leading-order PnP energy for…
PDF arXiv
arXiv cs.CVOct 30
Retrieval-Augmented Initialization with Dual-Guided Inversion for Flow-Based Inverse Problem Solvers
Hansam Cho, Seoung Bum Kim
Flow models have recently emerged as powerful generative priors for solving inverse problems. Despite their success, most existing flow-based inverse problem solvers initialize the reconstruction process with random Gaussian noise. However, recent studies in image generation have shown that the choice of initial noise can significantly influence generation outcomes. The role of initialization as a distinct design choice in flow-based inverse problem solving remains comparatively underexplored. In this work, we systematically evaluate noise initialization strategies while holding the downstream flow-based inverse solver fixed. We propose Retrieval-Augmented Initialization with Dual-Guided Inversion (RA-DGI), a method that constructs improved initial noise by leveraging retrieved proxy clean images and dual-guided inversion. Extensive experiments across multiple inverse problems demonstrate that the quality of the initial noise substantially affects reconstruction fidelity and perceptual performance. RA-DGI outperforms standard random initialization and achieves superior results compared to inference-time scaling strategies, while requiring less computational overhead. These…
PDF arXiv
arXiv eess.IVOct 30
Bayesian PET Reconstruction with Learned Flow Matching Priors via Langevin Sampling
Hengjia Ran, Huafeng Liu, Bo Zhao
Image reconstruction for positron emission tomography (PET) is a challenging inverse problem due to its inherent ill-posedness and Poisson counting noise, particularly in low-dose acquisitions. In this work, we introduce a novel Bayesian PET reconstruction framework that leverages a flow-matching generative prior learned from high-quality PET images. By incorporating this prior, we formulate PET reconstruction as a Bayesian inference problem and solve it via latent-space posterior sampling using the unadjusted Langevin algorithm. We evaluate the proposed method through numerical experiments and benchmark it against classical reconstruction methods and state-of-the-art deep learning-based baselines. The proposed method achieves improved reconstruction performance while enabling posterior uncertainty quantification.
PDF arXiv

Papers

Topics in this field

computed tomography · ct reconstruction · computed tomography cbct79 papers17 this week
diffusion priors · diffusion prior · imaging inverse77 papers12 this week

Recent papers

All papers in Research

arXiv cs.CVOct 31
3D Neural Rendering for Automated Ultrasonic Laser Scans
Thomas Pierson, Matthew Mitchell, Kira Perna, Anand Iyer, Eric Flynn +1
Laser doppler vibrometer (LDV) scans can be used to detect defects and cracks in manufactured parts before they are deployed. This work presents a method for reconstructing the steady-state surface wavefield of a three-dimensional part by combining automated robotic manipulation with neural rendering. A six-degree-of-freedom (6-DOF) robotic manipulator (UFactory xArm6) with an embedded piezoelectric transducer presented an acoustically excited test article to a stationary LDV at a sequence of known orientations, yielding a set of complex-valued 2D scans with poses derived analytically from the arm kinematics rather than from image-based pose estimation. These scans were used to train a novel complex-nerfacto model, a modified Neural Radiance Field (NeRF) pipeline that replaces the standard three-channel RGB output head with a two-channel output. Complex-nerfacto represents the real and imaginary velocity components of a wavefield, preserving the complex measurement without quantization. Reconstructions of a welded aluminum cylinder and a welded T-bracket were evaluated against measured scans using Z-normalized cross-correlation (ZNCC). The complex-valued pipeline achieved a mean…
PDF arXiv
arXiv cs.CVOct 30
Learning Clean Generative Processes from Noisy Measurements
Fabian Kruse, Lorenzo Baldassari, Ivan Dokmanić
We study how to learn a generative model of clean images using only noisy observations. Diffusion processes and their generalizations to non-Gaussian noise require posterior predictions at every noise level. We first show that it is possible to learn these predictions from independent noisy views using Noise2Noise regression, but only at or above the noise level of the training data. To learn the predictions below that level, needed to generate clean images, we impose consistency: a prediction should equal its expectation after continuing the sampling process. We derive conditions under which consistency uniquely determines the continuation to the clean image distribution and verify them for Poisson, binomial, Gamma, and negative-binomial measurements. Experiments on CIFAR-10, photon-counting microscopy, and synthetic aperture radar show that the required posterior predictions can be learned from noisy views alone and that the consistency relation repairs the continuation beyond the noise level of the data, which severely deteriorates without consistency.
PDF arXiv
arXiv eess.IVOct 30
Does Vendor Image Processing Matter? Learning Echocardiography Display Appearance from Native Beamspace Data
Elias Stenhede, Arian Ranjbar
Most echocardiography datasets contain vendor-processed images, but processing is often undocumented and may affect model performance and robustness. To address this, we train a small convolutional neural network to reproduce General Electric display appearance from native beamspace brightness-mode ultrasound, using 8 406 recordings from 605 unique patients. The beamspace CNN reduces test mean absolute error by 43% over the optimal pointwise brightness map, outperforms an equivalent Cartesian CNN and runs in real time on a CPU. Across the EchoXFlow and CAMUS datasets, unfiltered training followed by filtered evaluation yields the lowest mean Dice for every class, suggesting filtering may be useful as augmentation when evaluation data are filtered or their processing is unknown. Code and pretrained weights: https://github.com/Ahus-AIM/EchoAppearance
PDF arXiv
arXiv stat.MLOct 30
The Prior-Force Family of Implicit Denoiser Priors and Its Decisions in Linear Inverse Problems
Weijie Liang, Chuling Wen, Jian Lu
Implicit priors given by Gaussian denoisers successfully regularize a wide range of inverse problems in approaches such as plug-and-play (PnP), regularization by denoising (RED) and score-based methods. Such a prior depends on the noise level, and how it selects among signals consistent with the data remains largely uncharacterized. To put implicit and explicit priors on the same footing, we define the prior-force family $W_t=t^{-1}(D_t^{-1}-I)$ of any Gaussian denoiser $D_t$ at noise variance $t>0$ (inverse in the graph sense), the normalized denoising residual as an operator on clean signals. At a common effective prior weight, the equilibrium equations differ only in their prior terms: PnP uses $W_t$, RED/score-based methods the score force (the normalized residual as an operator on noisy inputs), and maximum a posteriori (MAP) estimation the MAP force (the negative log-prior gradient). For the minimum mean-square error (MMSE) denoisers of a prior $π$, we compare the corresponding energies at three levels. Without observation, raising the noise level favors PnP minima that retain more posterior variance. On signals consistent with the data, the leading-order PnP energy for…
PDF arXiv
arXiv cs.CVOct 30
Retrieval-Augmented Initialization with Dual-Guided Inversion for Flow-Based Inverse Problem Solvers
Hansam Cho, Seoung Bum Kim
Flow models have recently emerged as powerful generative priors for solving inverse problems. Despite their success, most existing flow-based inverse problem solvers initialize the reconstruction process with random Gaussian noise. However, recent studies in image generation have shown that the choice of initial noise can significantly influence generation outcomes. The role of initialization as a distinct design choice in flow-based inverse problem solving remains comparatively underexplored. In this work, we systematically evaluate noise initialization strategies while holding the downstream flow-based inverse solver fixed. We propose Retrieval-Augmented Initialization with Dual-Guided Inversion (RA-DGI), a method that constructs improved initial noise by leveraging retrieved proxy clean images and dual-guided inversion. Extensive experiments across multiple inverse problems demonstrate that the quality of the initial noise substantially affects reconstruction fidelity and perceptual performance. RA-DGI outperforms standard random initialization and achieves superior results compared to inference-time scaling strategies, while requiring less computational overhead. These…
PDF arXiv
arXiv eess.IVOct 30
Bayesian PET Reconstruction with Learned Flow Matching Priors via Langevin Sampling
Hengjia Ran, Huafeng Liu, Bo Zhao
Image reconstruction for positron emission tomography (PET) is a challenging inverse problem due to its inherent ill-posedness and Poisson counting noise, particularly in low-dose acquisitions. In this work, we introduce a novel Bayesian PET reconstruction framework that leverages a flow-matching generative prior learned from high-quality PET images. By incorporating this prior, we formulate PET reconstruction as a Bayesian inference problem and solve it via latent-space posterior sampling using the unadjusted Langevin algorithm. We evaluate the proposed method through numerical experiments and benchmark it against classical reconstruction methods and state-of-the-art deep learning-based baselines. The proposed method achieves improved reconstruction performance while enabling posterior uncertainty quantification.
PDF arXiv

Active contributors · 90 days

Activity · 12 weeks

Fields · AI · updated 30m ago

Diffusion priors

Found by the topic model: diffusion priors, computed tomography.

29papers this week+81.2% vs last week

All papers · 156 With code · 0

Newest Most relevant

90 papers this month

Open problems in Diffusion priors

Balancing initialization and stochasticity in diffusion-based inverse problems1 paper raises it · 1 addresses it
Excessive noise during sampling can erase gains from good initialization, degrading reconstruction quality.
- FLINT: Coupling Proximal Initialization and Bounded Stochasticity for Flow-Matching Inverse Problemsraises · addresses
Joint image-measurement consistency in inverse problems1 paper raises it · 1 addresses it
Most methods condition on noisy measurements but do not enforce consistency between reconstructed image and denoised measurement.
- Dual Flow: Joint Image-Measurement Flow for Inverse Problemsraises · addresses
Handling sparse, asynchronous, or missing sensor data in real-time reconstruction1 paper raises it · 1 addresses it
Real-world sensing systems often have irregular or incomplete measurements, limiting applicability of standard methods.
- Fault-tolerant shallow recurrent decoders for sparse sensing with missing sensor measurementsraises · addresses
Task-driven measurement design for single-pixel imaging1 paper raises it · 1 addresses it
Fixed illumination patterns may not be optimal for classification tasks, wasting acquisition time and photons.
- LIPS-PC: Learning Illumination Patterns for Single-Pixel Classificationraises · addresses
Generalization sensitivity to injected noise in diffusion sampling1 paper raises it · 1 addresses it
Small changes in noise during sampling can break strong generalization across models, affecting reproducibility.
- The role of injected noise in diffusion model generalizationraises · addresses

Named by an AI from these papers' abstracts.