Pith. sign in

REVIEW 4 major objections 5 minor 2 cited by

X-GRM: Large Gaussian Reconstruction Model for Sparse-view X-rays to Computed Tomography

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read X-GRM is a large feedforward Transformer that reconstructs a full 3D CT volume from six to ten X-ray projections in about one second, using fixed-position voxel-based Gaussians to make the volume both extractable and differentiable.

desk verdict Competent large-model sparse-view CT paper with a genuinely useful fixed-center Gaussian representation, but the clinical claim runs ahead of the synthetic-only evidence. read the letter →

arxiv 2505.15235 v2 pith:RRSKNCUO submitted 2025-05-21 eess.IV cs.CV

Yifan Liu , Wuyang Li , Weihao Yu , Chenxin Li , Alexandre Alahi , Max Meng , Yixuan Yuan This is my paper
classification eess.IVcs.CV
keywords computedtomographysparse-viewCTreconstructionX-rayimaging3DGaussianSplattingfeedforwardtransformervolumerenderingnovelviewsynthesismedical
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

X-GRM is a large feedforward Transformer that takes six to ten 2D X-ray projections of a body region and outputs a full 3D CT volume in about one second. The paper's central claim is that a scalable cross-view transformer paired with a new volume representation, Voxel-based Gaussian Splatting, gives better reconstruction quality and far faster inference than existing sparse-view CT methods, including per-sample optimization approaches that take minutes to hours. If that claim holds, sparse-view CT could shift from iterative, time-consuming reconstruction to instant learned reconstruction from many fewer projections, lowering radiation exposure and enabling time-sensitive uses. The experiments support the claim on a large curated set of 14,972 CT volumes with synthetic X-ray projections, plus cross-dataset and novel-view tests.

What carries the argument

The load-bearing object is Voxel-based Gaussian Splatting (VoxGS): a set of 3D Gaussians whose centers are locked to voxel centroids, each carrying only opacity $\alpha_i$, scale $s_i$, and rotation $r_i$. Locking positions lets the CT volume be extracted by direct indexing, $V(x,y,z)=\alpha_i$, with no trilinear interpolation, and dropping color is consistent with X-ray attenuation being a scalar line integral. The differentiable rasterizer for VoxGS supplies a rendering constraint during training, while the encoder-and-fusion ViT, with per-view patch tokens and all-to-all self-attention across views, supplies the capacity and cross-view reasoning that the paper argues prior CNN and voxel-grid models lack.

What would settle it

Run the trained X-GRM on a real cone-beam CT study with measured polyenergetic projections, or on simulated projections that add scatter and beam hardening, and compare its PSNR and SSIM against per-sample optimization methods; if the quality gap shrinks sharply or anatomical detail develops new artifacts, the paper's equivalence between 3DGS rasterization and X-ray imaging would be broken for practical CT.

Watch

Extended reading notes

Core claim

The paper proposes X-GRM, a one-pass model that maps sparse X-ray projections with their camera matrices to a voxelized density field $V \in \mathbb{R}^{M \times N \times L}$. Each projection is tokenized by a DINO-initialized ViT, given ray geometry through camera-ray-modulated adaptive layer norm, and all views are fused by a 16-layer all-to-all self-attention transformer. The fused tokens are decoded into Voxel-based Gaussian Splatting (VoxGS) attributes: every voxel center hosts a 3D Gaussian with opacity, scale, and rotation but no color, making CT extraction a direct opacity lookup and making X-ray rendering differentiable. The model is trained with a volume MSE loss plus a rendering loss combining L1 and D-SSIM, and it reports PSNR of 28.39, 28.86, and 29.21 dB for 6, 8, and 10 input views on the 680-volume test set, with SSIM of 0.873, 0.879, and 0.886. These numbers exceed the best feedforward baseline by roughly 3.6 to 3.8 dB while running about twice as fast, and exceed per-sample optimized methods by 4 to 5 dB while running hundreds to thousands of times faster. The same model also synthesizes unseen X-ray views with higher reported fidelity than NeRF- and 3DGS-based per-sample methods.

Load-bearing premise

The entire evaluation rests on synthetic X-ray projections rendered from CT volumes under simplified X-ray physics with no scatter or beam hardening, so the reported one-second reconstruction gains may not transfer to real clinical scanners if actual projection physics differ.

Editorial extensions

If this is right

  • With 6 to 10 input projections, a $256^3$ CT volume is reconstructed in about 0.9 seconds, a regime per-sample optimization methods cannot reach.
  • Because VoxGS supports differentiable X-ray rendering, the trained model can also synthesize novel projection views; the paper reports 49.44 dB PSNR on held-out views at 0.02 seconds per projection.
  • A single model trained with variable view counts (6, 8, or 10) serves different sparsity levels without retraining for each setting.
  • On unseen chest and pelvis datasets, the model retains a quality advantage over feedforward baselines and matches or beats per-sample optimization while being about 500 times faster, indicating out-of-distribution generalization.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the paper's explicit claims, VoxGS's opacity-as-density reading suggests the same fixed-lattice Gaussian head could be applied to other tomographic inverse problems whose forward operator is a line integral, such as PET or ultrasound computed tomography.
  • The paper's synthetic-only evaluation leaves an immediate stress test implicit: re-running the model on projections with beam hardening and scatter would quantify how much of the reported margin over per-sample optimization survives real scanner physics.
  • The model was trained and tested only with uniformly spaced views; an untested extension is non-uniform or limited-angle trajectories, where all-to-all cross-view attention may behave differently and missing angular coverage may expose the fixed voxel lattice's limits.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The manuscript proposes X-GRM, a feed-forward transformer-based model that reconstructs a 3D CT volume from 6, 8, or 10 sparse X-ray projections in about one second. The model uses a DINO-initialized ViT encoder to tokenize each projection, a fusion ViT to exchange information across views, and a novel Voxel-based Gaussian Splatting (VoxGS) representation in which isotropic Gaussians are placed at fixed voxel centers. Training combines an MSE loss on the extracted volume with L1 and D-SSIM losses on rendered X-rays, using sub-volume sampling to reduce memory. Experiments compare X-GRM with traditional, feedforward, and self-supervised baselines on a collected 14,972-volume dataset assembled from eight public CT datasets, and report consistent PSNR/SSIM gains, faster inference, and additional cross-dataset and novel-view-synthesis results.

Significance. If the reported results hold, X-GRM is a meaningful advance in sparse-view CT reconstruction: it combines a large-capacity transformer with a differentiable Gaussian volume representation, and the reported gains over strong baselines are consistent across 6/8/10-view settings. The paper contributes a sizable public-data-derived training set, a clean feed-forward formulation, and a promised code release, all of which are valuable for reproducibility and follow-up work. The main limitation is that every input X-ray in every experiment is synthesized with the TIGRE toolbox under a simplified monochromatic Beer-Lambert model; this makes the core technical contribution convincing as a proof of concept on synthetic data, but it does not by itself support the abstract's unqualified claims about 'various testing inputs' and clinical deployment.

major comments (4)
  1. [§4.1, §A.1, Abstract] All X-ray inputs across training, test, cross-dataset, and novel-view experiments are rendered from CT volumes with the TIGRE toolbox, and the equivalence between 3DGS rasterization and X-ray imaging is explicitly limited in A.1 to 'a simplified imaging model that accounts solely for isotropic absorption (per Beer-Lambert law).' The added Gaussian and Poisson noise in §4.1 does not model polyenergetic spectra, scatter, beam hardening, detector blur, or calibration offsets. As a result, the central claims in the Abstract and §1 that the model handles 'various testing inputs' and is suited to clinical workflows are stronger than the evidence supports. I would like to see the claims restricted to synthetic monochromatic projections, or ideally a validation on real paired X-ray/CT data (or at least a realistic polyenergetic scatter-inclusive simulation) to test physical domain shift.
  2. [§4.5, Table 6] The ablation study is described as performed on the 'ReconX-16K dataset,' but this dataset is never defined anywhere in the paper or appendix. Its source, number of volumes, split, resolution, and projection parameters are unknown, so the reader cannot determine whether the ablation is run on the same scale as the main experiments or whether the reported component rankings (e.g., 0.28 dB for pose, 0.55 dB for VoxGS, 0.52 dB for attention) are stable. This should be specified exactly, or the ablation should be moved to the main test split.
  3. [§4.2, Tables 2 and 3] The claim that X-GRM 'drastically outperforms' prior methods is based on single-run PSNR/SSIM numbers with no error bars, multiple seeds, or significance tests. Since feed-forward models are trained with stochastic optimization, run-to-run variance of several tenths of a dB is plausible at these resolutions, and some of the reported margins (e.g., 0.28 dB in Table 6a) are within that range. In addition, Table 2 reports timings on an A100 GPU while Table 3 uses an RTX 4090Ti, so the speed comparisons across tables are not directly comparable. Please report mean±std over at least three seeds and state the GPU configuration for each timing measurement.
  4. [§4.3, Table 4] The cross-dataset experiments on FUMPE and PENGWIN demonstrate generalization to unseen anatomies, but because the projections are still generated with the same TIGRE rendering pipeline, they do not demonstrate generalization to new acquisition physics. The text in §4.3 and the Abstract's phrase 'out-domain X-ray projections' suggest a broader domain shift than the experiment actually tests. Please rephrase these claims as anatomical-domain generalization, or add an experiment with a different forward model to support physical-domain generalization.
minor comments (5)
  1. [References, Table 3] Table 3 cites R2-Gaussian as [76], but the reference list places R2-Gaussian at [75]; moreover, the same work appears to be duplicated as references [74] and [75]. Please reconcile the numbering and deduplicate.
  2. [§4.4 vs. §A.3] The main text says novel-view synthesis is evaluated on 30 distinct CT samples, while §A.3 and Table 8 describe the 'sampled test set (40 samples)'. Please clarify which number is correct.
  3. [§3.5, Eq. (10)] The rendering loss weights λ_L1 and λ_SSIM are not reported. Please give their values, as well as the sub-volume sampling factor, so that the training objective is fully reproducible.
  4. [References, §3.3] Reference [78] is the object-detection DINO paper, but the text says the encoder is initialized from DINO pre-trained weights, which normally refers to the self-supervised ViT-DINO of Caron et al. Please correct the citation.
  5. [§3.5] The notation for sub-volume sampling uses K for both the number of views and the depth dimension (M/4×N/4×K/4), while the volume is earlier defined as M×N×L. Please use consistent dimensional notation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the CT prediction is learned against ground-truth volumes and external baselines, and the Beer-Lambert/3DGS equivalence is a standard physical model, not a self-referential construction.

full rationale

X-GRM's derivation chain is self-contained and non-circular. The forward model is defined by the Beer-Lambert line integral (Sec. 3.1), and the VoxGS representation fixes Gaussian positions at voxel centers and predicts opacities from fused X-ray tokens (Eqs. 5-8). Training minimizes a volume MSE against ground-truth CT volumes (Eq. 9) plus a rendering loss against ground-truth X-ray projections (Eq. 10). The predicted CT volume is therefore a supervised output of a learned mapping f_theta, not a repackaged version of the input projections. The relation V(x,y,z)=alpha_i is a definition of the chosen representation, not a derived claim that reduces to its inputs. No fitted parameter is renamed as a prediction: the reported PSNR/SSIM numbers are comparisons against FDK, SART, ASD-POCS, FBPConvNet, FreeSeed, DIF-Net, DIF-Gaussian, NAF, SAX-NeRF, and R2-Gaussian on held-out test volumes, with feedforward baselines retrained on the same data. Citations to X-LRM, R2-Gaussian, and DINO are used for dataset construction, rasterizers, and initialization; none of these are self-citations by the present authors, and none is invoked as an unverified uniqueness theorem. The main caveat, correctly identified as an external-validity concern rather than circularity, is that all X-ray projections, including cross-dataset tests, are synthesized with the TIGRE toolbox under the simplified monochromatic Beer-Lambert model (Sec. 4.1, A.1); this affects realism and clinical transfer but does not make any equation equivalent to its own input. The A.4 limitations about memory and very sparse inputs are practical constraints, not circular steps. Overall, no prediction in the paper is forced by construction or by self-citation.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The scientific claim is empirical and rests on standard deep learning plus three domain assumptions: (1) the Beer-Lambert/3DGS equivalence (Sec. 3.1, A.1), (2) the V=alpha_i volume extraction mapping (Eq. 8), and (3) TIGRE synthetic X-rays being representative of clinical acquisitions (Sec. 4.1). The only hand-chosen numbers that affect training are the rendering loss weights and the sub-volume sampling factor; the network weights are learned, not fitted constants.

free parameters (2)
  • Rendering loss weights lambda_L1 and lambda_SSIM = not reported
    Appear in Eq. 10; hand-chosen hyperparameters are never specified, making exact reproduction impossible.
  • Sub-volume sampling factor = 1/4 per dimension
    Training renders only M/4 x N/4 x K/4 sub-volumes to fit GPU memory (Sec. 3.5); this is a hand-chosen trade-off between memory and gradient fidelity.
assumptions (3)
  • domain assumption X-ray attenuation follows the Beer-Lambert law with monochromatic, scatter-free imaging; 3DGS rasterization is mathematically equivalent to X-ray projection under this model.
    Stated in Sec. 3.1 and A.1; this justifies using differentiable splatting for X-ray rendering and training, but ignores polyenergetic spectra, scatter, and beam hardening present in real CT.
  • domain assumption The CT volume density at a voxel can be read directly from the opacity alpha_i of the Gaussian fixed at that voxel center (Eq. 8).
    Assumes negligible contribution from neighboring Gaussians at the voxel center; this is not exact for overlapping Gaussians but is validated empirically by the ablations.
  • domain assumption Synthetic X-ray projections generated with TIGRE, with Gaussian and Poisson noise, are a sufficient proxy for clinical X-ray projections.
    The entire training and test pipeline uses simulated radiographs; generalization to real acquisitions is not demonstrated.

how reviews work

0 comments
Cite this review

Pith. "Pith review of X-GRM: Large Gaussian Reconstruction Model for Sparse-view X-rays to Computed Tomography." pith.science (2026). https://pith.science/paper/RRSKNCUO

@misc{pith2026250515235,
  author       = {Pith},
  title        = {Pith review of: X-GRM: Large Gaussian Reconstruction Model for Sparse-view X-rays to Computed Tomography},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RRSKNCUO}},
  note         = {Machine review of arXiv:2505.15235}
}
read the original abstract

Computed Tomography serves as an indispensable tool in clinical workflows, providing non-invasive visualization of internal anatomical structures. Existing CT reconstruction works are limited to small-capacity model architecture and inflexible volume representation. In this work, we present X-GRM (X-ray Gaussian Reconstruction Model), a large feedforward model for reconstructing 3D CT volumes from sparse-view 2D X-ray projections. X-GRM employs a scalable transformer-based architecture to encode sparse-view X-ray inputs, where tokens from different views are integrated efficiently. Then, these tokens are decoded into a novel volume representation, named Voxel-based Gaussian Splatting (VoxGS), which enables efficient CT volume extraction and differentiable X-ray rendering. This combination of a high-capacity model and flexible volume representation, empowers our model to produce high-quality reconstructions from various testing inputs, including in-domain and out-domain X-ray projections. Our codes are available at: https://github.com/CUHK-AIM-Group/X-GRM.

Figures

Figures reproduced from arXiv: 2505.15235 by the authors.

Figure 1
Figure 1. Our method achieves state-of-the-art reconstruction quality while maintaining the fastest [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. X-GRM is a large feed-forward transformer trained on a curated large CT reconstruction dataset. (a) X-ray Reconstruction Transformer efficiently encodes and fuses tokens from multiple X-ray projections, and (b) Voxel-based Gaussian Splatting enables both the efficient CT volume extraction and differentiable X-ray rendering. information (Sec. 3.3). Then, we discuss the design of VoxGS and how to regress Gaussian attr… view at source ↗
Figure 3
Figure 3. Qualitative comparison with traditional and feedforward methods. Results shown are from [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Qualitative comparison with self-supervised models. Results shown are from the test set [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Qualitative results of novel views [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: Additional qualitative comparison of novel view synthesis. [PITH_FULL_IMAGE:figures/full_fig_p016_6.png]
Figure 7
Figure 7. Figure 7: Qualitative comparison of reconstructed slices with traditional and feedforward models. [PITH_FULL_IMAGE:figures/full_fig_p017_7.png]
Figure 8
Figure 8. Figure 8: Qualitative comparison of reconstructed slices with self-supervised models. [PITH_FULL_IMAGE:figures/full_fig_p017_8.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. ILV: Iterative Latent Volumes for Fast and Accurate Sparse-View CT Reconstruction

    cs.CV 2026-03 conditional novelty 6.0 of 10

    ILV recovers fine anatomical detail in sparse-view CBCT by iteratively updating an explicit 3D latent volume with multi-view X-ray features and a learned prior, outperforming prior feed-forward and optimization method...

  2. VoxDet: Rethinking 3D Semantic Occupancy Prediction as Dense Object Detection

    cs.GR 2025-06 conditional novelty 6.0 of 10

    VoxDet reformulates 3D semantic occupancy prediction as dense object detection by deriving instance-boundary offsets from voxel class labels, and reports new state-of-the-art results on camera and LiDAR benchmarks.

Reference graph

Works this paper leans on

78 extracted references · 49 canonical work pages · cited by 2 Pith papers

  1. [1]

    Simultaneous algebraic reconstruction technique (sart): a superior implementation of the art algorithm

    Anders H Andersen and Avinash C Kak. Simultaneous algebraic reconstruction technique (sart): a superior implementation of the art algorithm. Ultrasonic imaging, 1984

    work page 1984

  2. [2]

    Lose the views: Limited angle ct reconstruction via implicit sinogram completion

    Rushil Anirudh, Hyojin Kim, Jayaraman J Thiagarajan, K Aditya Mohan, Kyle Champley, and Timo Bremer. Lose the views: Limited angle ct reconstruction via implicit sinogram completion. In CVPR, 2018

    work page 2018

  3. [3]

    Tigre: a matlab-gpu toolbox for cbct image reconstruction

    Ander Biguri, Manjit Dosanjh, Steven Hancock, and Manuchehr Soleimani. Tigre: a matlab-gpu toolbox for cbct image reconstruction. Biomedical Physics & Engineering Express, 2016

    work page 2016

  4. [4]

    Segmenting the inferior alveolar canal in cbcts volumes: the toothfairy challenge

    Federico Bolelli, Luca Lumetti, Shankeeth Vinayahalingam, Mattia Di Bartolomeo, Arrigo Pellacani, Kevin Marchesini, Niels Van Nistelrooij, Pieter Van Lierop, Tong Xi, Yusheng Liu, et al. Segmenting the inferior alveolar canal in cbcts volumes: the toothfairy challenge. IEEE Transactions on Medical Imaging, 2024

    work page 2024

  5. [5]

    Radiative gaussian splatting for efficient x-ray novel view synthesis

    Yuanhao Cai, Yixun Liang, Jiahao Wang, Angtian Wang, Yulun Zhang, Xiaokang Yang, Zongwei Zhou, and Alan Yuille. Radiative gaussian splatting for efficient x-ray novel view synthesis. In ECCV, 2024

    work page 2024

  6. [6]

    Structure-aware sparse-view x-ray 3d reconstruction

    Yuanhao Cai, Jiahao Wang, Alan Yuille, Zongwei Zhou, and Angtian Wang. Structure-aware sparse-view x-ray 3d reconstruction. In CVPR, 2024

    work page 2024

  7. [7]

    Baking gaussian splatting into diffusion denoiser for fast and scalable single-stage image-to-3d generation

    Yuanhao Cai, He Zhang, Kai Zhang, Yixun Liang, Mengwei Ren, Fujun Luan, Qing Liu, Soo Ye Kim, Jianming Zhang, Zhifei Zhang, et al. Baking gaussian splatting into diffusion denoiser for fast and scalable single-stage image-to-3d generation. arXiv preprint arXiv:2411.14384, 2024

    arXiv 2024

  8. [8]

    pixelsplat: 3d gaussian splats from image pairs for scalable generalizable 3d reconstruction

    David Charatan, Sizhe Lester Li, Andrea Tagliasacchi, and Vincent Sitzmann. pixelsplat: 3d gaussian splats from image pairs for scalable generalizable 3d reconstruction. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 19457–19467, 2024

    work page 2024

Show all 78 references
  1. [9]

    Mvsplat: Efficient 3d gaussian splatting from sparse multi-view images

    Yuedong Chen, Haofei Xu, Chuanxia Zheng, Bohan Zhuang, Marc Pollefeys, Andreas Geiger, Tat-Jen Cham, and Jianfei Cai. Mvsplat: Efficient 3d gaussian splatting from sparse multi-view images. In European Conference on Computer Vision, pages 370–386. Springer, 2024

    2024

  2. [10]

    Decomposed diffusion sampler for accelerating large-scale inverse problems

    Hyungjin Chung, Suhyeon Lee, and Jong Chul Ye. Decomposed diffusion sampler for accelerating large-scale inverse problems. In ICLR, 2024

    2024

  3. [11]

    Solving 3d inverse problems using pre-trained 2d diffusion models

    Hyungjin Chung, Dohoon Ryu, Michael T McCann, Marc L Klasky, and Jong Chul Ye. Solving 3d inverse problems using pre-trained 2d diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22542–22551, 2023

    2023

  4. [12]

    Representation of a function by its line integrals, with some radiological applications

    Allan Macleod Cormack. Representation of a function by its line integrals, with some radiological applications. Journal of applied physics, 1963

    1963

  5. [13]

    Objaverse-xl: A universe of 10m+ 3d objects

    Matt Deitke, Ruoshi Liu, Matthew Wallingford, Huong Ngo, Oscar Michel, Aditya Kusupati, Alan Fan, Christian Laforte, Vikram V oleti, Samir Yitzhak Gadre, et al. Objaverse-xl: A universe of 10m+ 3d objects. In NeurIPS, 2023

    2023

  6. [14]

    Objaverse: A universe of annotated 3d objects

    Matt Deitke, Dustin Schwenk, Jordi Salvador, Luca Weihs, Oscar Michel, Eli VanderBilt, Ludwig Schmidt, Kiana Ehsani, Aniruddha Kembhavi, and Ali Farhadi. Objaverse: A universe of annotated 3d objects. In CVPR, 2023

    2023

  7. [15]

    An image is worth 16x16 words: Transformers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby. An image is worth 16x16 words: Transformers for image recognition at...

    2021

  8. [16]

    Practical cone-beam algorithm

    Lee A Feldkamp, Lloyd C Davis, and James W Kress. Practical cone-beam algorithm. Josa a, 1984

    1984

  9. [17]

    Ddgs-ct: Direction-disentangled gaussian splatting for realistic volume rendering

    Zhongpai Gao, Benjamin Planche, Meng Zheng, Xiao Chen, Terrence Chen, and Ziyan Wu. Ddgs-ct: Direction-disentangled gaussian splatting for realistic volume rendering. In The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024. 10

    2024

  10. [18]

    Deep learning-based sinogram completion for low-dose ct

    Muhammad Usman Ghani and W Clem Karl. Deep learning-based sinogram completion for low-dose ct. In 2018 IEEE 13th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP), 2018

    2018

  11. [19]

    Lucidfusion: Generating 3d gaussians with arbitrary unposed images

    Hao He, Yixun Liang, Luozhou Wang, Yuanhao Cai, Xinli Xu, Hao-Xiang Guo, Xiang Wen, and Ying- cong Chen. Lucidfusion: Generating 3d gaussians with arbitrary unposed images. arXiv preprint arXiv:2410.15636, 2024

    2024 arXiv

  12. [20]

    Rsna 2023 abdominal trauma ai challenge: Review and outcomes

    Sebastiaan Hermans, Zixuan Hu, Robyn L Ball, Hui Ming Lin, Luciano M Prevedello, Ferco H Berger, Ibrahim Yusuf, Jeffrey D Rudie, Maryam Vazirabad, Adam E Flanders, et al. Rsna 2023 abdominal trauma ai challenge: Review and outcomes. Radiology: Artificial Intelligence, 2024

    2023

  13. [21]

    Lrm: Large reconstruction model for single image to 3d

    Yicong Hong, Kai Zhang, Jiuxiang Gu, Sai Bi, Yang Zhou, Difan Liu, Feng Liu, Kalyan Sunkavalli, Trung Bui, and Hao Tan. Lrm: Large reconstruction model for single image to 3d. In ICLR, 2024

    2024

  14. [22]

    Computerized transverse axial scanning (tomography): Part 1

    Godfrey N Hounsfield. Computerized transverse axial scanning (tomography): Part 1. description of system. The British journal of radiology, 1973

    1973

  15. [23]

    Computed medical imaging

    Godfrey N Hounsfield. Computed medical imaging. Science, 1980

    1980

  16. [24]

    Amos: A large-scale abdominal multi-organ benchmark for versatile medical image segmentation

    Yuanfeng Ji, Haotian Bai, Chongjian Ge, Jie Yang, Ye Zhu, Ruimao Zhang, Zhen Li, Lingyan Zhanng, Wanling Ma, Xiang Wan, et al. Amos: A large-scale abdominal multi-organ benchmark for versatile medical image segmentation. In NeurIPS, 2022

    2022

  17. [25]

    Deep convolutional neural network for inverse problems in imaging

    Kyong Hwan Jin, Michael T McCann, Emmanuel Froustey, and Michael Unser. Deep convolutional neural network for inverse problems in imaging. TIP, 2017

    2017

  18. [26]

    Deep convolutional neural network for inverse problems in imaging

    Kyong Hwan Jin, Michael T McCann, Emmanuel Froustey, and Michael Unser. Deep convolutional neural network for inverse problems in imaging. IEEE transactions on image processing, 2017

    2017

  19. [27]

    Deep-learning-assisted detection and segmentation of rib fractures from ct scans: Development and validation of fracnet

    Liang Jin, Jiancheng Yang, Kaiming Kuang, Bingbing Ni, Yiyi Gao, Yingli Sun, Pan Gao, Weiling Ma, Mingyu Tan, Hui Kang, Jiajun Chen, and Ming Li. Deep-learning-assisted detection and segmentation of rib fractures from ct scans: Development and validation of fracnet. eBioMedicine, 2020

    2020

  20. [28]

    3d gaussian splatting for real-time radiance field rendering

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimkühler, and George Drettakis. 3d gaussian splatting for real-time radiance field rendering. ACM Trans. Graph., 2023

    2023

  21. [29]

    Improving 3d imaging with pre-trained perpendicular 2d diffusion models

    Suhyeon Lee, Hyungjin Chung, Minyoung Park, Jonghyuk Park, Wi-Sun Ryu, and Jong Chul Ye. Improving 3d imaging with pre-trained perpendicular 2d diffusion models. In ICCV, 2023

    2023

  22. [30]

    Grounding image matching in 3d with mast3r

    Vincent Leroy, Yohann Cabon, and Jérôme Revaud. Grounding image matching in 3d with mast3r. In European Conference on Computer Vision, pages 71–91. Springer, 2024

    2024

  23. [31]

    Instant3d: Fast text-to-3d with sparse-view generation and large reconstruction model

    Jiahao Li, Hao Tan, Kai Zhang, Zexiang Xu, Fujun Luan, Yinghao Xu, Yicong Hong, Kalyan Sunkavalli, Greg Shakhnarovich, and Sai Bi. Instant3d: Fast text-to-3d with sparse-view generation and large reconstruction model. In ICLR, 2024

    2024

  24. [32]

    Abdomenatlas: A large-scale, detailed-annotated, & multi-center dataset for efficient transfer learning and open algorithmic benchmarking

    Wenxuan Li, Chongyu Qu, Xiaoxi Chen, Pedro RAS Bassi, Yijia Shi, Yuxiang Lai, Qian Yu, Huimin Xue, Yixiong Chen, Xiaorui Lin, et al. Abdomenatlas: A large-scale, detailed-annotated, & multi-center dataset for efficient transfer learning and open algorithmic benchmarking. Medic...

    2024

  25. [33]

    3dgr-ct: Sparse-view ct reconstruction with a 3d gaussian representation

    Yingtai Li, Xueming Fu, Han Li, Shang Zhao, Ruiyang Jin, and S Kevin Zhou. 3dgr-ct: Sparse-view ct reconstruction with a 3d gaussian representation. Medical Image Analysis, page 103585, 2025

    2025

  26. [34]

    Learning deep intensity field for extremely sparse-view cbct reconstruction

    Yiqun Lin, Zhongjin Luo, Wei Zhao, and Xiaomeng Li. Learning deep intensity field for extremely sparse-view cbct reconstruction. In MICCAI, 2023

    2023

  27. [35]

    Learning 3d gaussians for extremely sparse-view cone-beam ct reconstruction

    Yiqun Lin, Hualiang Wang, Jixiang Chen, and Xiaomeng Li. Learning 3d gaussians for extremely sparse-view cone-beam ct reconstruction. In MICCAI, 2024

    2024

  28. [36]

    Deepsparse: A foundation model for sparse-view cbct reconstruction

    Yiqun Lin, Hualiang Wang, Jixiang Chen, Jiewen Yang, Jiarong Guo, and Xiaomeng Li. Deepsparse: A foundation model for sparse-view cbct reconstruction. arXiv preprint arXiv:2505.02628, 2025

    2025

  29. [37]

    Cˆ 2rv: Cross- regional and cross-view learning for sparse-view cbct reconstruction

    Yiqun Lin, Jiewen Yang, Hualiang Wang, Xinpeng Ding, Wei Zhao, and Xiaomeng Li. Cˆ 2rv: Cross- regional and cross-view learning for sparse-view cbct reconstruction. In CVPR, 2024

    2024

  30. [38]

    Zero- 1-to-3: Zero-shot one image to 3d object

    Ruoshi Liu, Rundi Wu, Basile Van Hoorick, Pavel Tokmakov, Sergey Zakharov, and Carl V ondrick. Zero- 1-to-3: Zero-shot one image to 3d object. In Proceedings of the IEEE/CVF international conference on computer vision, pages 9298–9309, 2023. 11

    2023

  31. [39]

    Pelvic fracture segmentation using a multi-scale distance-weighted neural network

    Yanzhen Liu, Sutuke Yibulayimu, Yudi Sang, Gang Zhu, Yu Wang, Chunpeng Zhao, and Xinbao Wu. Pelvic fracture segmentation using a multi-scale distance-weighted neural network. In MICCAI, 2023

    2023

  32. [40]

    Sgdr: Stochastic gradient descent with warm restarts

    Ilya Loshchilov and Frank Hutter. Sgdr: Stochastic gradient descent with warm restarts. arXiv preprint arXiv:1608.03983, 2016

    2016 arXiv

  33. [41]

    Decoupled weight decay regularization

    Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101, 2017

    2017 arXiv

  34. [42]

    Freeseed: Frequency-band- aware and self-guided network for sparse-view ct reconstruction

    Chenglong Ma, Zilong Li, Junping Zhang, Yi Zhang, and Hongming Shan. Freeseed: Frequency-band- aware and self-guided network for sparse-view ct reconstruction. In MICCAI, 2023

    2023

  35. [43]

    A new dataset of computed-tomography angiography images for computer-aided detection of pulmonary embolism

    Mojtaba Masoudi, Hamid-Reza Pourreza, Mahdi Saadatmand-Tarzjan, Noushin Eftekhari, Fateme Shafiee Zargar, and Masoud Pezeshki Rad. A new dataset of computed-tomography angiography images for computer-aided detection of pulmonary embolism. Scientific Data, 2018

    2018

  36. [44]

    Mediastinal lesion analysis (mela) dataset, 2022

    MELA Challenge Organizers. Mediastinal lesion analysis (mela) dataset, 2022

    2022

  37. [45]

    Pytorch: An imperative style, high-performance deep learning library

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. Pytorch: An imperative style, high-performance deep learning library. In NeurIPS, 2019

    2019

  38. [46]

    Scalable diffusion models with transformers

    William Peebles and Saining Xie. Scalable diffusion models with transformers. In Proceedings of the IEEE/CVF international conference on computer vision, pages 4195–4205, 2023

    2023

  39. [47]

    A local update strategy for iterative reconstruction from projections

    Ken Sauer and Charles Bouman. A local update strategy for iterative reconstruction from projections. IEEE Transactions on Signal Processing, 1993

    1993

  40. [48]

    Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the luna16 challenge

    Arnaud Arindra Adiyoso Setio, Alberto Traverso, Thomas De Bel, Moira SN Berens, Cas Van Den Bogaard, Piergiorgio Cerello, Hao Chen, Qi Dou, Maria Evelina Fantacci, Bram Geurts, et al. Validation, comparison, and combination of algorithms for automatic detection of pulmonary no...

    2017

  41. [49]

    Nerp: implicit neural representation learning with prior embedding for sparsely sampled image reconstruction

    Liyue Shen, John Pauly, and Lei Xing. Nerp: implicit neural representation learning with prior embedding for sparsely sampled image reconstruction. IEEE Transactions on Neural Networks and Learning Systems, 2022

    2022

  42. [51]

    Image reconstruction in circular cone-beam computed tomography by constrained, total-variation minimization

    Emil Y Sidky and Xiaochuan Pan. Image reconstruction in circular cone-beam computed tomography by constrained, total-variation minimization. Physics in Medicine & Biology, 2008

    2008

  43. [52]

    Hisplat: Hierar- chical 3d gaussian splatting for generalizable sparse-view reconstruction

    Shengji Tang, Weicai Ye, Peng Ye, Weihao Lin, Yang Zhou, Tao Chen, and Wanli Ouyang. Hisplat: Hierar- chical 3d gaussian splatting for generalizable sparse-view reconstruction. arXiv preprint arXiv:2410.06245, 2024

    2024 arXiv

  44. [53]

    Mv-dust3r+: Single-stage scene reconstruction from sparse views in 2 seconds

    Zhenggang Tang, Yuchen Fan, Dilin Wang, Hongyu Xu, Rakesh Ranjan, Alexander Schwing, and Zhicheng Yan. Mv-dust3r+: Single-stage scene reconstruction from sparse views in 2 seconds. arXiv preprint arXiv:2412.06974, 2024

    2024 arXiv

  45. [54]

    Triposr: Fast 3d object reconstruction from a single image

    Dmitry Tochilkin, David Pankratz, Zexiang Liu, Zixuan Huang, Adam Letts, Yangguang Li, Ding Liang, Christian Laforte, Varun Jampani, and Yan-Pei Cao. Triposr: Fast 3d object reconstruction from a single image. arXiv preprint arXiv:2403.02151, 2024

    2024 arXiv

  46. [55]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems, 30, 2017

    2017

  47. [56]

    A non-exponential transmittance model for volumetric scene representations

    Delio Vicini, Wenzel Jakob, and Anton Kaplanyan. A non-exponential transmittance model for volumetric scene representations. ACM Transactions on Graphics (TOG), 40(4):1–16, 2021

    2021

  48. [57]

    Dudotrans: dual-domain transformer for sparse-view ct reconstruction

    Ce Wang, Kun Shang, Haimiao Zhang, Qian Li, and S Kevin Zhou. Dudotrans: dual-domain transformer for sparse-view ct reconstruction. In International Workshop on Machine Learning for Medical Image Reconstruction, 2022

    2022

  49. [58]

    Vggt: Visual geometry grounded transformer

    Jianyuan Wang, Minghao Chen, Nikita Karaev, Andrea Vedaldi, Christian Rupprecht, and David Novotny. Vggt: Visual geometry grounded transformer. arXiv preprint arXiv:2503.11651, 2025. 12

    2025 arXiv

  50. [59]

    Pixel2mesh: Generating 3d mesh models from single rgb images

    Nanyang Wang, Yinda Zhang, Zhuwen Li, Yanwei Fu, Wei Liu, and Yu-Gang Jiang. Pixel2mesh: Generating 3d mesh models from single rgb images. In ECCV, 2018

    2018

  51. [60]

    Pf-lrm: Pose-free large reconstruction model for joint pose and shape prediction

    Peng Wang, Hao Tan, Sai Bi, Yinghao Xu, Fujun Luan, Kalyan Sunkavalli, Wenping Wang, Zexiang Xu, and Kai Zhang. Pf-lrm: Pose-free large reconstruction model for joint pose and shape prediction. arXiv preprint arXiv:2311.12024, 2023

    2023 arXiv

  52. [61]

    Continuous 3d perception model with persistent state

    Qianqian Wang, Yifei Zhang, Aleksander Holynski, Alexei A Efros, and Angjoo Kanazawa. Continuous 3d perception model with persistent state. arXiv preprint arXiv:2501.12387, 2025

    2025 arXiv

  53. [62]

    Dust3r: Geometric 3d vision made easy

    Shuzhe Wang, Vincent Leroy, Yohann Cabon, Boris Chidlovskii, and Jerome Revaud. Dust3r: Geometric 3d vision made easy. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 20697–20709, 2024

    2024

  54. [63]

    Sts-tooth: A multi-modal dental dataset for semi- supervised deep learning image segmentation, September 2024

    yaqi wang. Sts-tooth: A multi-modal dental dataset for semi- supervised deep learning image segmentation, September 2024

    2024

  55. [64]

    Freesplat: Generalizable 3d gaussian splatting towards free view synthesis of indoor scenes.Advances in Neural Information Processing Systems, 37:107326–107349, 2024

    Yunsong Wang, Tianxin Huang, Hanlin Chen, and Gim Hee Lee. Freesplat: Generalizable 3d gaussian splatting towards free view synthesis of indoor scenes.Advances in Neural Information Processing Systems, 37:107326–107349, 2024

    2024

  56. [65]

    Image quality assessment: from error visibility to structural similarity

    Zhou Wang, Alan C Bovik, Hamid R Sheikh, and Eero P Simoncell. Image quality assessment: from error visibility to structural similarity. TIP, 2004

    2004

  57. [66]

    Meshlrm: Large reconstruction model for high-quality meshes

    Xinyue Wei, Kai Zhang, Sai Bi, Hao Tan, Fujun Luan, Valentin Deschaintre, Kalyan Sunkavalli, Hao Su, and Zexiang Xu. Meshlrm: Large reconstruction model for high-quality meshes. arXiv preprint arXiv:2404.12385, 2024

    2024 arXiv

  58. [67]

    Pq-net: A generative part seq2seq network for 3d shapes

    Rundi Wu, Yixin Zhuang, Kai Xu, Hao Zhang, and Baoquan Chen. Pq-net: A generative part seq2seq network for 3d shapes. In CVPR, 2020

    2020

  59. [68]

    Grm: Large gaussian reconstruction model for efficient 3d reconstruction and generation

    Yinghao Xu, Zifan Shi, Wang Yifan, Hansheng Chen, Ceyuan Yang, Sida Peng, Yujun Shen, and Gordon Wetzstein. Grm: Large gaussian reconstruction model for efficient 3d reconstruction and generation. In European Conference on Computer Vision, pages 1–20. Springer, 2024

    2024

  60. [69]

    Deep rib fracture instance segmentation and classification from ct on the ribfrac challenge

    Jiancheng Yang, Rui Shi, Liang Jin, Xiaoyang Huang, Kaiming Kuang, Donglai Wei, Shixuan Gu, Jianying Liu, Pengfei Liu, Zhizhong Chai, Yongjie Xiao, Hao Chen, Liming Xu, Bang Du, Xiangyi Yan, Hao Tang, Adam Alessio, Gregory Holste, Jiapeng Zhang, Xiaoming Wang, Jianye He, Lixua...

    2024

  61. [70]

    Fast3r: Towards 3d reconstruction of 1000+ images in one forward pass

    Jianing Yang, Alexander Sax, Kevin J Liang, Mikael Henaff, Hao Tang, Ang Cao, Joyce Chai, Franziska Meier, and Matt Feiszli. Fast3r: Towards 3d reconstruction of 1000+ images in one forward pass. arXiv preprint arXiv:2501.13928, 2025

    2025 arXiv

  62. [71]

    No pose, no problem: Surprisingly simple 3d gaussian splats from sparse unposed images

    Botao Ye, Sifei Liu, Haofei Xu, Xueting Li, Marc Pollefeys, Ming-Hsuan Yang, and Songyou Peng. No pose, no problem: Surprisingly simple 3d gaussian splats from sparse unposed images. arXiv preprint arXiv:2410.24207, 2024

    2024 arXiv

  63. [72]

    pixelnerf: Neural radiance fields from one or few images

    Alex Yu, Vickie Ye, Matthew Tancik, and Angjoo Kanazawa. pixelnerf: Neural radiance fields from one or few images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 4578–4587, 2021

    2021

  64. [73]

    Region of interest reconstruction from truncated data in circular cone-beam ct

    Lifeng Yu, Yu Zou, Emil Y Sidky, Charles A Pelizzari, Peter Munro, and Xiaochuan Pan. Region of interest reconstruction from truncated data in circular cone-beam ct. TMI, 2006

    2006

  65. [75]

    R2-gaussian: Rectifying radiative gaussian splatting for tomographic reconstruction

    Ruyi Zha, Tao Jun Lin, Yuanhao Cai, Jiwen Cao, Yanhao Zhang, and Hongdong Li. R2-gaussian: Rectifying radiative gaussian splatting for tomographic reconstruction. In NeurIPS, 2024

    2024

  66. [76]

    Naf: Neural attenuation fields for sparse-view cbct reconstruc- tion

    Ruyi Zha, Yanhao Zhang, and Hongdong Li. Naf: Neural attenuation fields for sparse-view cbct reconstruc- tion. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2022: 25th International Conference, Singapore, September 18–22, 2022, Proceedings, Part VI, page...

    2022

  67. [77]

    X-lrm: X-ray large reconstruction model for extremely sparse-view computed tomography recovery in one second

    Guofeng Zhang, Ruyi Zha, Hao He, Yixun Liang, Alan Yuille, Hongdong Li, and Yuanhao Cai. X-lrm: X-ray large reconstruction model for extremely sparse-view computed tomography recovery in one second. arXiv preprint arXiv:2503.06382, 2025. 13

    2025

  68. [78]

    Dino: Detr with improved denoising anchor boxes for end-to-end object detection

    Hao Zhang, Feng Li, Shilong Liu, Lei Zhang, Hang Su, Jun Zhu, Lionel M Ni, and Heung-Yeung Shum. Dino: Detr with improved denoising anchor boxes for end-to-end object detection. arXiv preprint arXiv:2203.03605, 2022

    2022 arXiv

  69. [79]

    Gs-lrm: Large reconstruction model for 3d gaussian splatting

    Kai Zhang, Sai Bi, Hao Tan, Yuanbo Xiangli, Nanxuan Zhao, Kalyan Sunkavalli, and Zexiang Xu. Gs-lrm: Large reconstruction model for 3d gaussian splatting. In ECCV, 2024

    2024

  70. [80]

    Flare: Feed-forward geometry, appearance and camera estimation from uncalibrated sparse views

    Shangzhan Zhang, Jianyuan Wang, Yinghao Xu, Nan Xue, Christian Rupprecht, Xiaowei Zhou, Yujun Shen, and Gordon Wetzstein. Flare: Feed-forward geometry, appearance and camera estimation from uncalibrated sparse views. arXiv preprint arXiv:2502.12138, 2025. 14 A Technical Append...

    2025

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.