Training-free Linear Image Inverses via Flows
read the original abstract
Solving inverse problems without any training involves using a pretrained generative model and making appropriate modifications to the generation process to avoid finetuning of the generative model. While recent methods have explored the use of diffusion models, they still require the manual tuning of many hyperparameters for different inverse problems. In this work, we propose a training-free method for solving linear inverse problems by using pretrained flow models, leveraging the simplicity and efficiency of Flow Matching models, using theoretically-justified weighting schemes, and thereby significantly reducing the amount of manual tuning. In particular, we draw inspiration from two main sources: adopting prior gradient correction methods to the flow regime, and a solver scheme based on conditional Optimal Transport paths. As pretrained diffusion models are widely accessible, we also show how to practically adapt diffusion models for our method. Empirically, our approach requires no problem-specific tuning across an extensive suite of noisy linear inverse problems on high-dimensional datasets, ImageNet-64/128 and AFHQ-256, and we observe that our flow-based method for solving inverse problems improves upon closely-related diffusion-based methods in most settings.
This paper has not been read by Pith yet.
Forward citations
Cited by 8 Pith papers
-
FlowADMM: Plug-and-play ADMM with Flow-based Renoise-Denoise Priors
FlowADMM replaces stochastic renoise-denoise steps in flow-based plug-and-play methods with a deterministic expectation operator inside ADMM, yielding convergence guarantees under weak Lipschitz conditions and state-o...
-
Assistron: Bayesian Shared Autonomy with Off-the-shelf Vision-Language-Action Models
Assistron combines pre-trained VLA models with phase-aware Bayesian shared autonomy and flow matching guidance to raise task success rates and lower human workload in manipulation benchmarks without model fine-tuning.
-
Frequency-Guided Action Diffusion via Sub-Frequency Manifold Traversal
FGO guides diffusion policy generation via expanding spectral bands on sub-frequency manifolds to improve action smoothness on 15 robotic manipulation tasks.
-
Saving Foundation Flow-Matching Priors for Inverse Problems
FMPlug adapts foundation flow-matching models into practical priors for inverse problems by combining instance-guided warm-start with sharp Gaussianity regularization, showing superior results on image restoration and...
-
Real-Time Execution of Action Chunking Flow Policies
Real-time chunking (RTC) allows diffusion- and flow-based action chunking policies to execute smoothly and asynchronously, maintaining high success rates on dynamic tasks even with significant inference latency.
-
Smoother Action Chunking Flow Policy via Prior-Corrected Orthogonal Trust-Region Guidance
POTR augments RTC guidance for flow-matching policies by adding a data-prior scale to the weight schedule and constraining the perpendicular component of the guidance vector within a trust region, yielding smoother ac...
-
Understanding Asynchronous Inference Methods for Vision-Language-Action Models
Controlled benchmarks show per-step residual correction (A2C2) as most effective for VLA asynchronous inference up to d=8 delays on Kinetix with over 90% solve rate, outperforming inpainting and conditioning while tra...
-
Flow Matching Guide and Code
Flow Matching is a generative modeling framework with mathematical foundations, design choices, extensions, and open-source PyTorch code for applications like image and text generation.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.