Efficient Unrolled Networks for Large-Scale 3D Inverse Problems

• URL: http://arxiv.org/abs/2601.02141v1
• Date: Mon, 05 Jan 2026 14:12:43 GMT
• Title: Efficient Unrolled Networks for Large-Scale 3D Inverse Problems
• Abstract summary: In large-scale problems, such as 3D imaging, most existing methods fail to incorporate the operator in the architecture due to the prohibitive amount of memory required by global forward operators.<n>We present a domain partitioning strategy and normal operator approximations that enable the training of end-to-end reconstruction models incorporating forward operators into their architecture.<n>The proposed method achieves state-of-the-art performance on 3D X-ray cone-beam tomography and 3D multi-coil accelerated MRI, while requiring only a single GPU for both training and inference.
• Score: 11.979276488739364
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Deep learning-based methods have revolutionized the field of imaging inverse problems, yielding state-of-the-art performance across various imaging domains. The best performing networks incorporate the imaging operator within the network architecture, typically in the form of deep unrolling. However, in large-scale problems, such as 3D imaging, most existing methods fail to incorporate the operator in the architecture due to the prohibitive amount of memory required by global forward operators, which hinder typical patching strategies. In this work, we present a domain partitioning strategy and normal operator approximations that enable the training of end-to-end reconstruction models incorporating forward operators of arbitrarily large problems into their architecture. The proposed method achieves state-of-the-art performance on 3D X-ray cone-beam tomography and 3D multi-coil accelerated MRI, while requiring only a single GPU for both training and inference.

Related papers

• iLRM: An Iterative Large 3D Reconstruction Model [34.69940725179257]
  We introduce an iterative Large 3D Reconstruction Model (iLRM) that generates 3D Gaussian representations through an iterative refinement mechanism.<n> Experimental results on widely used datasets, such as RE10K and DL3DV, demonstrate that iLRM outperforms existing methods in both reconstruction quality and speed.
  arXiv  Detail & Related papers  (2025-07-31T06:33:07Z)
• Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass [68.78222900840132]
  We propose Fast 3D Reconstruction (Fast3R), a novel multi-view generalization to DUSt3R that achieves efficient and scalable 3D reconstruction by processing many views in parallel.<n>Fast3R demonstrates state-of-the-art performance, with significant improvements in inference speed and reduced error accumulation.
  arXiv  Detail & Related papers  (2025-01-23T18:59:55Z)
• Radiant: Large-scale 3D Gaussian Rendering based on Hierarchical Framework [13.583584930991847]
  We propose Radiant, a hierarchical 3DGS algorithm designed for large-scale scene reconstruction.<n>We show that Radiant improved reconstruction quality by up to 25.7% and reduced up to 79.6% end-to-end latency.
  arXiv  Detail & Related papers  (2024-12-07T05:48:00Z)
• Large Spatial Model: End-to-end Unposed Images to Semantic 3D [79.94479633598102]
  Large Spatial Model (LSM) processes unposed RGB images directly into semantic radiance fields. LSM simultaneously estimates geometry, appearance, and semantics in a single feed-forward operation. It can generate versatile label maps by interacting with language at novel viewpoints.
  arXiv  Detail & Related papers  (2024-10-24T17:54:42Z)
• Deep Learning-based Cross-modal Reconstruction of Vehicle Target from Sparse 3D SAR Image [6.499547636078961]
  We introduce cross-modal learning and propose a Cross-Modal 3D-SAR Reconstruction Network (CMAR-Net) for enhancing sparse 3D SAR images of vehicle targets by fusing optical information.<n>CMAR-Net achieves efficient training and reconstructs sparse 3D SAR images, which are derived from highly sparse-aspect observations, into visually structured 3D vehicle images.
  arXiv  Detail & Related papers  (2024-06-06T15:18:59Z)
• Plug-and-Play Regularization on Magnitude with Deep Priors for 3D Near-Field MIMO Imaging [0.0]
  Near-field radar imaging systems are used in a wide range of applications such as concealed weapon detection and medical diagnosis. We consider the problem of the three-dimensional (3D) complex-valued reflectivity by enforcing regularization on its magnitude.
  arXiv  Detail & Related papers  (2023-12-26T12:25:09Z)
• HyperNeRFGAN: Hypernetwork approach to 3D NeRF GAN [0.4606926019499777]
  This paper introduces HyperNeRFGAN, a Generative Adversarial Network (GAN) architecture employing a hypernetwork paradigm to transform a Gaussian noise into the weights of a NeRF architecture. Despite its notable simplicity in comparison to existing state-of-the-art alternatives, the proposed model demonstrates superior performance on a diverse range of image datasets.
  arXiv  Detail & Related papers  (2023-01-27T10:21:18Z)
• GLEAM: Greedy Learning for Large-Scale Accelerated MRI Reconstruction [50.248694764703714]
  Unrolled neural networks have recently achieved state-of-the-art accelerated MRI reconstruction. These networks unroll iterative optimization algorithms by alternating between physics-based consistency and neural-network based regularization. We propose Greedy LEarning for Accelerated MRI reconstruction, an efficient training strategy for high-dimensional imaging settings.
  arXiv  Detail & Related papers  (2022-07-18T06:01:29Z)
• Operator Sketching for Deep Unrolling Networks [5.025654873456756]
  We propose a new paradigm for designing efficient deep unrolling networks using operator sketching. Our numerical experiments on X-ray CT image reconstruction demonstrate the effectiveness of sketched unrolling schemes.
  arXiv  Detail & Related papers  (2022-03-21T17:34:18Z)
• Accurate and Lightweight Image Super-Resolution with Model-Guided Deep Unfolding Network [63.69237156340457]
  We present and advocate an explainable approach toward SISR named model-guided deep unfolding network (MoG-DUN) MoG-DUN is accurate (producing fewer aliasing artifacts), computationally efficient (with reduced model parameters), and versatile (capable of handling multiple degradations) The superiority of the proposed MoG-DUN method to existing state-of-theart image methods including RCAN, SRDNF, and SRFBN is substantiated by extensive experiments on several popular datasets and various degradation scenarios.
  arXiv  Detail & Related papers  (2020-09-14T08:23:37Z)
• Hierarchical Amortized Training for Memory-efficient High Resolution 3D GAN [52.851990439671475]
  We propose a novel end-to-end GAN architecture that can generate high-resolution 3D images. We achieve this goal by using different configurations between training and inference. Experiments on 3D thorax CT and brain MRI demonstrate that our approach outperforms state of the art in image generation.
  arXiv  Detail & Related papers  (2020-08-05T02:33:04Z)
• Learning Deformable Image Registration from Optimization: Perspective, Modules, Bilevel Training and Beyond [62.730497582218284]
  We develop a new deep learning based framework to optimize a diffeomorphic model via multi-scale propagation. We conduct two groups of image registration experiments on 3D volume datasets including image-to-atlas registration on brain MRI data and image-to-image registration on liver CT data.
  arXiv  Detail & Related papers  (2020-04-30T03:23:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.