Radiative Gaussian Splatting for Efficient X-ray Novel View Synthesis

arXiv:2403.04116 · eess.IV, cs.CV · Submitted 2024-03-07 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Electrical Engineering. The day in electrical engineering: signals, systems and the hardware that carries them.

Rosa: I'm Rosa.

Dev: And I'm Dev, here with Taro.

Rosa: We're looking at the paper: "Radiative Gaussian Splatting for Efficient X-ray Novel View Synthesis".

Title and authors: Rosa: So, this paper, "Radiative Gaussian Splatting for Efficient X-ray Novel View Synthesis," it's basically about taking three dee Gaussian splatting and making it work much better for X-rays <ref:2403.04116#pg1,Radiative Gaussian Splatting for Efficient X-ray Novel View Synthesis>.

Dev: Right. It tackles the slow training and inference speed problems that methods like NeRF have with X-ray novel view synthesis, which is a real bottleneck right now.

Taro: I mean, if you're trying to do this kind of reconstruction on real scanner data, those long training times and slow speeds are just not practical for anything outside of a controlled lab setting.

Rosa: Exactly. The authors propose X-Gaussian, which is this three dee Gaussian splatting framework built around two main things: a new way to model the Gaussian point cloud and a new way to render it <ref:2403.04116#pg1>.

Dev: Let's unpack that model first. They redesign the radiative Gaussian point cloud model by using something called a Radiation Intensity Response Function, or RIRF, instead of the Spherical Harmonics, or SH, that you usually see in standard three dee Gaussian splatting <ref:2403.04116#pg1,radiative Gaussian point cloud model>.

Taro: What does excluding the influence of view direction mean for X-rays specifically?

Rosa: Well, X-ray imaging is naturally isotropic—it doesn't care which way you look at it. So they designed their RIRF to exclude that view direction entirely when learning how to predict the radiation intensity of a three dee point <ref:2403.04116#pg1,to predict the radiation intensity of>.

Dev: That’s key because in original methods, you often have to account for the view direction, which makes things computationally expensive during both training and rendering.

Taro: So by removing that dependency, they're trying to simplify the physics model so it matches how an X-ray machine actually works.

Rosa: Right. And then they pair that with this Differentiable Radiative Rasterization, or DRR, which they implemented using a CUDA kernel for speed on the GPU.

Dev: The DRR is their rendering pipeline. They've made it so that projecting those three dee Gaussians onto the 2D detector plane doesn't have to do all that heavy computation related to view direction in both the forward and backward passes, which speeds things up significantly <ref:2403.04116#pg1>.

Title and authors: Taro: I wonder how much faster this actually is compared to standard volume rendering or ray tracing methods?

Rosa: The experiments show that X-Gaussian is seventy-three times faster for inference and only requires about fifteen percent of the training time compared to other state-of-the-art methods, which is a big deal.

Dev: That speed improvement, combined with the model changes, leads to over six point five decibels better performance than existing methods. That's a solid lift in quality for the same computational cost.

Taro: It’s interesting that they also mentioned that their X-Gaussian can improve sparse-view CT reconstruction performance by yielding PSNR improvements of fifteen point one nine, thirteen point zero one, and thirteen point five three dB compared to methods like FDK or SART.

Rosa: That shows it's not just about rendering novel views; it has practical value in things like reconstructing CT scans where you only get a few projections.

Dev: Moving on to the specifics of how they got this setup running, they customized an Angle-pose Cuboid Uniform Initialization, or ACUI, strategy.

Taro: What does that initialization strategy actually do without needing the heavy structure-from-motion algorithm?

Rosa: The ACUI strategy uses the actual parameters of the X-ray scanner to calculate the intrinsic and extrinsic matrices for you directly. Then they set up a cuboid that completely encloses the scanned object, and they uniformly sample points inside that cuboid to initialize where those Gaussian point clouds are centered.

Dev: So it avoids running Structure-from-Motion, which is usually a nightmare for X-ray data because of issues like low contrast and overlapping layers, which is a huge time saver.

Taro: It sounds like they're bypassing the need to solve a massive geometric problem upfront just to get the training started.

Rosa: Right. And as an engineer focused on real-time performance, I want to know about the failure modes. What happens when things go wrong in this setup?

Title and authors: Dev: The paper mentions that they found that for optimal balance between performance and speed, having a number of features Nf equal to sixteen seems to be the sweet spot.

Taro: So, if we push beyond sixteen points, does the system start slowing down or losing quality faster than expected?

Rosa: They also found that the best performance, memory cost, and fastest inference speed all happen when you use an initialization interval of eight. That’s a concrete parameter to tune before you run a full training cycle.

Dev: It shows they've really mapped out the operational parameters for this framework, moving beyond just theoretical concepts to something you can actually set up on a loop.

Taro: So, in short, they took the isotropic nature of X-rays and built a whole pipeline around it—model, rendering, and initialization—that significantly cuts down the computational load compared to standard neural radiance fields.

Rosa: That’s the gist of it for now. We're going to wrap up this discussion on "Radiative Gaussian Splatting for Efficient X-ray Novel View Synthesis."

Dev: This work shows how tailoring a three deeGS framework specifically to the physics of X-ray imaging, using that RIRF model and the DRR kernel, can lead to substantial gains in speed and quality for novel view synthesis <ref:2403.04116#pg1>.

Taro: I just think it’s promising because it tackles the fundamental speed issues inherent in these types of reconstruction tasks by changing how we model the intensity response.

Rosa: We've covered what X-Gaussian is, why they changed the SH to RIRF, and how that translates into a faster rendering pipeline and better initialization using ACUI.

Dev: The key numbers are those performance gains: over six point five decibels better quality while achieving seventy-three times faster inference and only fifteen percent of the training time.

Taro: And the practical application in sparse-view CT reconstruction shows it's not just theoretical, it has real utility for medical imaging workflows.

Rosa: So, this paper is a solid step forward by creating a specialized three deeGS approach tailored for X-ray novel view synthesis, and we'll be looking at what comes next in the field soon <ref:2403.04116#pg1,X-ray novel view synthesis>.

The paper's summary: Rosa: So, to wrap up this paper, X-Gaussian is essentially taking three dee Gaussian splatting and making it work for X-ray novel view synthesis much faster than existing methods like NeRF.

Dev: Right. It’s all about redesigning the way the point cloud models light, which they call a radiative Gaussian point cloud model, using something called a radiation intensity response function or RIRF instead of spherical harmonics.

Rosa: So they're modeling how radiation intensity behaves in a way that ignores the view direction because X-rays are naturally isotropic. That’s the core idea there.

Dev: And then they pair that with a differentiable radiative rasterization method, which is implemented using CUDA for real speed on the GPU. They’ve made sure this rendering process doesn't have to deal with all that complex view direction math in both the training and rendering steps.

Rosa: So what does this mean for us? It means they are tackling the slow training and inference problems that plague current three deeGS methods specifically for medical imaging like X-rays.

Dev: The results show this system is over six point five decibels better in quality compared to state-of-the-art methods, while it’s only using about fifteen percent of the training time.

Rosa: That speed gain is huge; they say it’s seventy three times faster for inference and only needs a fraction of the training time.

Dev: Plus, they show it can improve sparse view CT reconstruction performance by fifteen point one nine, thirteen point zero one, and thirteen point five three decibels compared to other neural methods for things like FDK or SART.

Rosa: That’s significant because it shows this isn't just a theoretical rendering trick; it actually has practical value when you’re reconstructing CT scans where you only get a few projections.

Dev: The authors also showed that their initialization strategy, which they call angle-pose cuboid uniform initialization or ACUI, works really well without needing the heavy structure-from-motion algorithm.

Rosa: Using scanner parameters to compute the camera matrices and sampling points inside a cuboid instead of running SfM is a major time saver for starting any three dee reconstruction project.

Dev: So, while they’ve done some real work on making this pipeline efficient for X-rays, they also laid out some limitations in their ablation study.

Rosa: They found that to get the best balance of performance and speed, having a number of features equal to sixteen seems optimal, and an initialization interval of eight is the sweet spot for training.

Dev: That tells us we can start tuning this framework with those specific parameters in mind instead of just running it blindly.

Rosa: It sounds like they’ve really mapped out how to make three dee Gaussian splatting a viable tool for X-ray novel view synthesis by focusing on the physics of the imaging process itself.

Dev: This work suggests that tailoring three deeGS specifically to the properties of X-ray imaging, using that RIRF model and the DRR kernel, can lead to substantial gains in speed and quality for novel view synthesis tasks.

Rosa: We’ve covered what X-Gaussian is, why they changed the SH to RIRF, and how that translates into a faster rendering pipeline and better initialization using ACUI.

Dev: The key numbers are those performance gains: over six point five decibels better quality while achieving seventy three times faster inference and only fifteen percent of the training time.

Rosa: And the practical application in sparse-view CT reconstruction shows it's not just theoretical, it has real utility for medical imaging workflows.

Dev: So, this paper is a solid step forward by creating a specialized three deeGS approach tailored for X-ray novel view synthesis, and we'll be looking at what comes next in the field soon.

The paper's improvements: Rosa: So, let's talk about the specific improvements X-Gaussian suggests for this framework beyond just being faster on training.

Dev: Right, like they are pushing a few more things to make it work better in real scenarios where things aren't perfectly controlled lab setups.

Rosa: They focus on the model itself, specifically how they handle the intensity response. They emphasize using that radiation intensity response function or RIRF instead of spherical harmonics.

Dev: That’s because the RIRF is designed to inherently ignore the view direction, which matches how X-ray imaging actually works and simplifies things for our rendering pipeline.

Rosa: And they also made a specific tweak to the way they initialize those Gaussian point clouds with this angle-pose cuboid uniform initialization strategy.

Dev: So instead of relying on the slow structure-from-motion algorithm, which we know is tough on X-ray data, they use scanner parameters to set up camera matrices and then just sample points in a box around the object.

Rosa: That means you can get started much faster without needing that heavy geometry processing first.

Dev: And they also identified some specific settings for their system that give them the best performance, memory cost, and inference speed. They found that having sixteen features is a sweet spot for the number of features, and an initialization interval of eight is what works best.

Rosa: That gives us concrete numbers to test when we try to run this outside of a perfect simulation environment.

Dev: Exactly. This shows they’ve done more than just tweak the math; they’ve mapped out operational parameters for this method, which is important for setting up any kind of real-time loop.

Rosa: So, what does all this mean for someone who is trying to use this in a field roboticist's setup?

Dev: It means that if you can get the hardware calibrated using those scanner parameters and run the initialization with those specific feature counts, you’re going to see much better results than just throwing it at a standard NeRF model.

Rosa: The implication is that for novel view synthesis in X-ray, we can move away from these super slow training cycles and get something that actually runs quickly enough for practical applications.

Dev: It points toward a future where three deeGS methods can be adapted more easily to imaging modalities that have unique physical properties, like the isotropic nature of X-rays.

Rosa: It's interesting because we see this work alongside papers on robust transmission design and antenna optimization, which suggests this efficiency gain is being looked at across different hardware challenges.

Dev: Yeah, so even when dealing with hardware distortions or complex signal processing, the goal seems to be achieving a much more efficient representation of the three dee scene.

Conclusion: Rosa: So, to wrap up, X-Gaussian is essentially a three dee Gaussian splatting framework specifically tailored for X-ray novel view synthesis that significantly cuts down on training time and speeds up inference.

Dev: Right, it uses the RIRF model and the DRR rendering kernel to make sure we aren't stuck with those slow, heavy computations related to view direction in both the forward and backward passes.

Rosa: And they’ve shown that by using this angle-pose cuboid uniform initialization strategy instead of running structure-from-motion, we can save a massive amount of time on getting started.

Dev: That initialization trick is pretty smart because it bypasses the need forSfM altogether, which is usually a huge bottleneck for X-ray data because of low contrast and overlapping layers.

Rosa: The big numbers are pretty clear: over six point five decibels better quality while enjoying seventy three times faster inference and only requiring about fifteen percent of the training time.

Dev: That speed improvement combined with the model changes leads to over six point five decibels better performance than existing methods, which is a solid lift in quality for the same computational cost.

Rosa: Plus, they also showed that it can improve sparse view CT reconstruction performance by fifteen point one nine, thirteen point zero one, and thirteen point five three decibels compared to other neural methods for things like FDK or SART.

Dev: That shows it’s not just about rendering novel views; it has practical value when you’re reconstructing CT scans where you only get a few projections.

Rosa: So, this paper is a solid step forward by creating a specialized three deeGS approach tailored for X-ray novel view synthesis, and we'll be looking at what comes next in the field soon.

Dev: This work suggests that tailoring three deeGS specifically to the physics of X-ray imaging, using that RIRF model and the DRR kernel, can lead to substantial gains in speed and quality for novel view synthesis tasks.

Rosa: We’ve covered what X-Gaussian is, why they changed the SH to RIRF, and how that translates into a faster rendering pipeline and better initialization using ACUI.

Dev: The key numbers are those performance gains: over six point five decibels better quality while achieving seventy three times faster inference and only fifteen percent of the training time.

Rosa: And the practical application in sparse-view CT reconstruction shows it's not just theoretical, it has real utility for medical imaging workflows.

Dev: So, this paper is a solid step forward by creating a specialized three deeGS approach tailored for X-ray novel view synthesis, and we'll be looking at what comes next in the field soon.

Taro: I’m just interested in what happens when the world misbehaves with this system. If we try to run trials outside the lab, how long can it actually stay stable before things start going sideways?

Rosa: That’s a fair question, Taro. We haven't done long-term field trials on this one yet, but the authors did focus on those operational parameters like the number of features and initialization interval eight as tuning knobs.

Dev: Right. From a controls engineer standpoint, I want to know about the safety case for running this in real-time loops; what’s our latency tolerance when we use that DRR kernel?

Taro: So, if we push the feature count beyond sixteen or change those initialization parameters, how does the system react gracefully when it hits a failure mode? Does it crash or does it degrade predictably?

Rosa: The paper did mention that for optimal balance, those specific settings are best. It stops working when you go outside that sweet spot, but within those bounds, the performance is very stable and fast.

Dev: So the safety case is about sticking to those optimized parameters to avoid unpredictable behavior in the real-time loop environment.

Taro: It’s good to know it has defined boundaries for its operation. That helps us plan how we test its recovery when things get messy out there.

Johns Hopkins University · HKUST(GZ) · Shanghai Jiao Tong University

eess.IV, cs.CV

Submitted: 2024-03-07

Updated: 2026-11-17

Comments: ECCV 2024; The first 3D Gaussian Splatting-based method for X-ray 3D reconstruction

Code: https://github.com/caiyuanhao1998/X-Gaussian

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 86/100

The gist: X-Gaussian is a 3D Gaussian splatting-based framework proposed for X-ray novel view synthesis, which addresses the slow training and inference speed issues inherent in existing methods like NeRF by…

Key concepts

Radiative Gaussian Point Cloud Model
This is the core 3D representation of an object made up of many small, colored Gaussian blobs. Unlike standard models, it uses a Radiation Intensity Response Function (RIRF) instead of Spherical Harmonics. This RIRF predicts radiation intensity without needing to know the view direction, which mimics how X-rays behave.
Differentiable Radiative Rasterization (DRR)
This is the rendering pipeline that turns the 3D point cloud into a 2D image. DRR uses a CUDA implementation to project Gaussians onto the detector plane. Crucially, it avoids complex view-direction calculations in both training and inference, making it much faster than traditional methods.
Angle-pose Cuboid Uniform Initialization (ACUI)
This is a smart way to start the 3D scene setup without needing slow Structure-from-Motion (SfM). It uses known scanner parameters to calculate camera matrices and then uniformly places the initial Gaussian points inside a cuboid around the object, quickly setting up the model.
Radiation Intensity Response Function (RIRF)
The RIRF is a mathematical function used in this framework to predict how much radiation intensity a point emits. By using an RIRF, the model ignores view direction dependency. This simplification is key because X-ray imaging has an isotropic property, meaning the intensity looks the same regardless of which way you look.

Terminology

Summary

X-Gaussian is a 3D Gaussian splatting-based framework proposed for X-ray novel view synthesis, which addresses the slow training and inference speed issues inherent in existing methods like NeRF by proposing a radiative Gaussian point cloud model and an initialization strategy that avoids time-consuming SfM

The Core Proposal

The paper introduces X-Gaussian, a 3D Gaussian splatting-based framework for X-ray novel view synthesis. This framework is built upon two key techniques: redesigning a radiative Gaussian point cloud model inspired by the isotropic nature of X-ray imaging and developing a Differentiable Radiative Rasterization (DRR) with CUDA implementation. p. 2 The model replaces the Spherical Harmonics (SH) used in original 3DGS with a Radiation Intensity Response Function (RIRF) to predict radiation intensity, which excludes the influence of view direction, matching the isotropic property of X-ray imaging. p. 2

Initialization Strategy

The authors customize an Angle-pose Cuboid Uniform Initialization (ACUI) strategy for camera calibration parameters and Gaussian point clouds. This strategy exploits the parameters of the X-ray scanner to compute intrinsic and extrinsic matrices, followed by uniformly sampling 3D points within a cuboid enclosing the scanned object to initialize the center positions of the Gaussian point clouds. p. 3 This ACUI avoids running the Structure-from-Motion (SfM) algorithm, which is unsuitable for X-ray imaging due to issues like low contrast and overlapping layers. p. 1

Rendering Pipeline

The rendering process utilizes a Differentiable Radiative Rasterization (DRR) based on the radiative Gaussian point cloud model. The DRR involves projecting 3D Gaussians to the 2D detector plane, calculating the projected covariance matrix in camera coordinates, and blending ordered points to obtain the final intensity image. p. 5 This DRR avoids complex computations related to view direction in the forward and backward processes, leading to cheaper training costs and faster inference speed compared to RGB rasterization. p. 6

Performance Gains

Experiments demonstrate that X-Gaussian significantly outperforms state-of-the-art (SOTA) methods by over 6.5 dB while enjoying 73× faster inference speed and only requiring 15% of training time. p. 8 The method shows faster convergence, with the radiative Gaussian point clouds forming the basic shape of the object as early as the 1000-th iteration. p. 3 Furthermore, X-Gaussian can improve sparse-view CT reconstruction performance by yielding significant PSNR improvements compared to other NVS methods. p. 2

Ablation Study Insights

The break-down ablation study confirms the effectiveness of the proposed components. Replacing SfM with ACUI reduced training time by 34% while improving performance by 1. p. 266 dB. Removing the view-direction computation from rasterization further accelerated training and inference speed, reducing training time by 54. p. 610% and inference speed by 2. p. 01×. The analysis of the number of features Nf suggests that Nf = 16 achieves a more optimal balance between performance and speed. p. 4 The best performance, memory cost, and fastest inference speed are achieved at an initialization interval d = 8. p. 8

Conclusion

The paper concludes that X-Gaussian is the first 3DGS-based framework for X-ray novel view synthesis. It successfully proposes a radiative Gaussian point cloud model based on X-ray imaging properties and a differentiable radiative rasterization CUDA kernel. p. 2 The combination of the RIRF model and the ACUI initialization strategy results in superior performance, speed, and training efficiency compared to SOTA methods. p. 8 The method also shows practical value in sparse-view CT reconstruction tasks. p. 0

References

Andersen, A.H., Kak, A.C.: Simultaneous algebraic reconstruction technique (sart): a superior implementation of the art algorithm

Armato III, S.G., McLennan, G., Bidaut, L., McNitt-Gray, M.F., Meyer, C.R., Reeves, A.P., Zhao, B., Aberle, D.R., Henschke, C.I., Hoffman, E.A.: The lung image database consortium (lidc) and image database resource initiative (idri): a completed reference database of lung nodules on ct scans

Barron, J.T., Mildenhall, B., Tancik, M., Hedman, P., Martin-Brualla, R., Srinivasan, P.P.: Mip-nerf: A multiscale representation for anti-aliasing neural radiance fields

Barron, J.T., Mildenhall, B., Verbin, D., Srinivasan, P.P., Hedman, P.: Mip-nerf 360: Unbounded anti-aliased neural radiance fields

Biguri, A., Dosanjh, M., Hancock, S., Soleimani, M.: Tigre: a matlab-gpu toolbox for cbct image reconstruction

Boone, J., Shah, N., Nelson, T.: A comprehensive analysis of coefficients for pendant-geometry cone-beam breast computed tomography

Boone, J.M., Nelson, T.R., Lindfors, K.K., Seibert, J.A.: Dedicated breast ct: radiation dose and image quality evaluation

Cai, Y., Wang, J., Yuille, A. p. 0, Zhou, Z., Wang, A.: Structure-aware sparse-view x-ray 3d reconstruction

Chen, A., Xu, Z., Geiger, A., Yu, J., Su, H.: Tensorf: Tensorial radiance fields

Chen, B., Ning, R.: Cone-beam volume ct breast imaging: Feasibility study

Chen, Z., Funkhouser, T., Hedman, P., Tagliasacchi, A.: Mobilenerf: Exploiting the polygon rasterization pipeline for efficient neural field rendering on mobile architectures

Cormack, A.M.: Representation of a function by its line integrals, with some radiological applications

Cormack, A.M.: Representation of a function by its line integrals, with some radiological applications ii. p. 3 Journal of Applied Physics (1964) 1

Feldkamp, L.A., Davis, L.C., Kress, J.W.: Practical cone-beam algorithm

Guide, D.: Cuda c programming guide

Hounsfield, G.N.: Computerized transverse axial scanning (tomography): Part 1 description of system

Hounsfield, G.N.: Computed medical imaging

Hu, S., Liu, Z.: Gauhuman: Articulated gaussian splatting from monocular human videos

Hu, T., Liu, S., Chen, Y., Shen, T., Jia, J.: Efficientnerf efficient neural radiance fields

Li, R., Gao, H., Tancik, M., Kanazawa, A.: Nerfacc: Efficient sampling accelerates nerfs

Liang, Y., Yang, X., Lin, J., Li, H., Xu, X., Chen, Y.: Luciddreamer: Towards high-fidelity text-to-3d generation via interval score matching

Liang, Z., Zhang, Q. p. 0, Feng, Y., Shan, Y., Jia, K.: Gs-ir: 3d gaussian splatting for inverse rendering

Luiten, J., Kopanas, G., Leibe, B., Ramanan, D.: Neat: Neural adaptive tomography

Kopanas, G., Philip, J., Leimkühler, T., Drettakis, G.: Point-based neural rendering with per-view optimization

Matsuki, H., Murai, R., Kelly, P.H., Davison, A.J.: Gaussian splatting slam

Mildenhall, B., Srinivasan, P., Tancik, M., Barron, J., Ramamoorthi, R., Ng, R.: Nerf: Representing scenes as neural radiance fields for view synthesis

Müller, T., Evans, A., Schied, C., Keller, A.: Instant neural graphics primitives with a multiresolution hash encoding

Pan, J., Zhou, T., Han, Y., Jiang, M., et al.: Variable weighted ordered subset image reconstruction algorithm

Sauer, K., Bouman, C.: A local update strategy for iterative reconstruction from projections

Scarfe, W.C., Farman, A.G., Sukovic, P., et al.: Clinical applications of cone-beam computed tomography in dental practice

Sidky, E.Y., Pan, X.: Image reconstruction in circular cone-beam computed tomography by constrained, total-variation minimization

Snavely, N., Seitz, S.M., Szeliski, R.: Photo tourism: exploring photo collections in 3d

Verbin, D., Hedman, P., Mildenhall, B., Zickler, T., Barron, J.T., Srinivasan, P.P.: Bakedsdf: Meshing neural sdfs for real-time view synthesis

Zang, G., Idoughi, R., Li, R., Wonka, P., Heidrich, W.: Intratomo: self-supervised learning-based tomography via sinogram synthesis and prediction

Zbijewski, W., Defrise, M., Viergever, M.A., Beekman, F.J.: Statistical reconstruction for x-ray ct systems with non-continuous detectors

Zha, R., Zhang, Y. p. 0 p. 0, Li, H.: Naf: neural attenuation fields for sparse-view cbct reconstruction

Zwicker, M., Pfister, H., Van Baar, J., Gross, M.: Ewa volume splatting

Zha, R., Zhang, Y., Li, H.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Radiative Gaussian Point Cloud Model Improvement: Replace Spherical Harmonics (SH) used in original 3DGS with a Radiation Intensity Response Function (RIRF) that excludes the influence of view direction to model X-ray radiation intensity as isotropic.

  2. Differentiable Rasterization Enhancement: Develop a Differentiable Radiative Rasterization (DRR) with CUDA implementation that avoids complex computations related to the view direction in the forward and backward processes, leading to faster training costs and inference speed.

  3. Initialization Strategy Optimization: Implement the Angle-pose Cuboid Uniform Initialization (ACUI) strategy, which does not need to execute the SfM algorithm, by using scanner parameters to compute camera matrices and uniformly sampling points within a cuboid enclosing the object, thus significantly reducing training time.

  4. Performance Gains: The improved X-Gaussian system can outperform state-of-the-art methods by 6.5 dB while enjoying 73x inference speed and only requiring 15% of training time.

  5. Sparse-View CT Reconstruction Enhancement: The system can improve sparse-view CT reconstruction PSNR by 15.19, 13.01, and 13.53 dB compared to using standard NeRF methods for reconstruction tasks like FDK, SART, and ASD-POCS.

Sources

Related papers