Pith. sign in

REVIEW 4 major objections 5 minor 2 cited by

EventSplat: 3D Gaussian Splatting from Moving Event Cameras for Real-time Rendering

T0 review · 4 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read EventSplat trains 3D Gaussian Splatting from moving event cameras alone, reaching millisecond rendering at quality above event-based NeRFs.

desk verdict Real-time event-only 3DGS is a genuine, well-demonstrated advance, but the state-of-the-art quality claim overreaches the evidence: the main tables omit the event-based 3DGS baselines the paper itself cites. read the letter →

arxiv 2412.07293 v2 pith:WPQQERMI submitted 2024-12-10 cs.CV

Toshiya Yura , Ashkan Mirzaei , Igor Gilitschenski This is my paper · ORCID
classification cs.CV
keywords eventcameras3DGaussiansplattingnovelviewsynthesisevent-to-videoinitializationcubicsplineposeinterpolationreal-timerenderinglog-intensitydifferencehighdynamicrange
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

This paper claims that 3D Gaussian Splatting can be supervised directly by event camera data, without any RGB images, to reconstruct static scenes and render novel views in real time. The key move is to treat event streams as measurements of log-intensity change and to train the Gaussian representation by matching accumulated events to the log-difference of rendered views. The authors show this approach matches or beats existing event-based neural radiance fields in visual quality while rendering roughly an order of magnitude faster. This would matter because event cameras work where ordinary cameras blur or saturate, and real-time rendering makes event-only view synthesis practical.

What carries the argument

The central identity is the log-difference equation: over a time interval [a, b], the event camera's output approximates E(a, b) = integral of the time derivative of log intensity, which can be estimated twice—once by accumulating events into the image D, and once by rendering two views from the Gaussian scene and subtracting their log-remosaiced images to form Dhat. The training loop minimizes the discrepancy between these two approximations. Two supporting mechanisms carry the method: remosaicing of the rendered RGB image back into the Bayer pattern so that per-pixel event data can be compared against the correct color channel, and a two-stage initialization where a pretrained event-to-video model synthesizes images, structure-from-motion produces initial 3D points, and cubic spline interpolation supplies poses at arbitrary event times.

What would settle it

Take a real event sequence with dense ground-truth poses, then train the method repeatedly with the pose stream artificially downsampled (for example, to one-tenth of the original rate). If reconstruction quality stalls or collapses when interpolated poses deviate from the true trajectory, the high-rate-pose assumption is confirmed as load-bearing.

Watch

Extended reading notes

Core claim

On its own terms, the paper introduces a fully event-driven training pipeline for 3D Gaussian Splatting. For a random sub-trajectory of the event stream, it accumulates events into a difference image D that approximates the integral of log-intensity change, and it rasterizes the current set of 3D Gaussians at the two endpoint poses to produce a rendered log-difference image Dhat. The reconstruction loss between D and Dhat, applied only where events actually occurred, drives the Gaussian optimization. To make this work, the authors add two ingredients: an event-to-video model whose output is fed to structure-from-motion to initialize Gaussian positions, and cubic spline interpolation of the camera trajectory so that arbitrary event timestamps get accurate poses. The experiments report that on synthetic scenes the method achieves comparable PSNR and better SSIM and LPIPS than Robust-e-NeRF, and on real scenes it outperforms both Robust-e-NeRF and E2VID-plus-3DGS on all three metrics, with rendering times of a few milliseconds per frame.

Load-bearing premise

The method relies on the camera poses being sampled at a high constant rate so that cubic spline interpolation yields accurate poses for any event timestamp; if the pose stream is too coarse or drifts, the accumulated event image and the rendered log-difference image will be misaligned and training degrades.

Editorial extensions

If this is right

  • Event-only novel view synthesis becomes practical in real time, with rendering times in the single milliseconds instead of tens to hundreds of milliseconds.
  • The method is usable in fast-motion and high-dynamic-range conditions where conventional cameras suffer from motion blur or saturation, because events encode relative log-intensity changes.
  • The pretrained event-to-video prior plus structure-from-motion provides a viable initialization for 3D Gaussian Splatting without any RGB supervision, reducing the need for dense color images.
  • Because the supervision signal is a log-intensity difference, the reconstructed scene carries an unknown absolute intensity offset that can be corrected by a simple linear transformation at inference time.

Reading between the lines

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

  • A natural extension would be to treat the random sub-trajectory length as a curriculum, starting with short intervals for local geometry and lengthening them to capture global structure, which the current equal random sampling only approximates.
  • The log-difference supervision could be combined with joint optimization of the spline control points, potentially relaxing the assumption that constant-rate poses are accurate enough for interpolation.
  • The static-scene restriction suggests a direct follow-up: attach per-Gaussian motion or a continuous time axis to render dynamic scenes from event streams, which the paper explicitly leaves for future work.
  • The remosaicing step is a simple channel-wise masking that could be replaced by a learned or more accurate color-processing model, possibly improving color fidelity on real event cameras.
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 paper introduces EventSplat, a 3D Gaussian Splatting method trained directly from event camera streams for novel view synthesis. The training signal is a comparison between an accumulated event image D, computed by summing polarities over a randomly sampled sub-trajectory, and a rendered log-difference image D_hat, obtained by rasterizing the Gaussian scene at the start and end poses of that sub-trajectory, remosaicing the resulting color images, and taking the log difference. The method uses an event-to-video model to generate images that initialize Structure from Motion, and cubic spline interpolation to estimate camera poses at arbitrary event times. Experiments on seven synthetic scenes and five EDS real scenes compare against Robust-e-NeRF and E2VID+3DGS, reporting that EventSplat renders at 2.5–6 ms per frame, roughly two orders of magnitude faster than the NeRF baseline, with better mean SSIM and LPIPS on synthetic scenes and better PSNR, SSIM, and LPIPS on real scenes. Ablations show that event-to-video guided initialization and cubic spline interpolation each improve quality.

Significance. If validated, this work is significant: it is among the first to adapt 3D Gaussian Splatting to event-only supervision, and it demonstrates a clear and large rendering-speed advantage over event-based NeRF methods, an important practical bottleneck. The core speed result is robust and well supported by the reported timings. The ablation study supports the two main design choices. However, the broader claim of 'state-of-the-art quality' is not yet established, because the main experimental comparison omits concurrent event-based 3DGS baselines, and because the evaluation protocol appears to use test-time reference data for a linear color transform, which can inflate reported metrics. The paper is well written and the method is simple enough to reproduce, but the evaluation and the precision of the claims need strengthening before the headline conclusions can be accepted.

major comments (4)
  1. [Sec. 5.2, Tables 1–2, Appendix B.2] The primary evaluation compares only Robust-e-NeRF and E2VID+3DGS. The paper cites several event-based 3DGS methods (EV-GS, Event3DGS, E2GS) in Sec. 2.3, but the only direct comparison is Appendix Table 4, which reports mean PSNR and SSIM over just four grayscale synthetic scenes, with no per-scene values, no error bars, and no rendering-time comparison. Because the abstract and Sec. 1 claim state-of-the-art quality among event-only view synthesis methods, this omission is load-bearing. The authors should either include these baselines in the main tables with per-scene results and variance, or substantially weaken the claim to 'comparable quality at real-time speed.'
  2. [Sec. 5.3 and Sec. 5.6] The inference-time linear color transform is described as being 'designed to adjust our predictions' and later as requiring 'evaluation data as a reference.' If the scale and offset are optimized on the test views, the reported PSNR, SSIM, and LPIPS values are not a strict hold-out evaluation, because the test reference is used to fit a per-scene affine transform before computing the metrics. This can overstate reconstruction quality. Please clarify whether the transform is fit only on training views; if it uses test views, provide results without this fitting or with the transform estimated from training data only.
  3. [Table 1 and Sec. 5.4.1] The mean synthetic PSNR for EventSplat is 28.14 dB versus 28.19 dB for Robust-e-NeRF, so the abstract's blanket claim of 'higher visual fidelity' is not uniformly supported. The claim should be restricted to the metrics and settings where the improvement holds (SSIM and LPIPS on synthetic, all metrics on real scenes), and the PSNR deficit on synthetic scenes should be discussed, for example in terms of event accumulation noise, remosaicing, or the linear color transform.
  4. [Sec. 4.4 and Sec. 5.1] The method assumes constant-rate poses sampled at high frequency and uses cubic spline interpolation to assign poses at event times. The ablation in Table 3 shows a modest PSNR gain from adding cubic spline interpolation (18.75 to 18.86), but there is no sensitivity analysis to pose sampling rate or pose noise. Since misalignment between the accumulated event image D and the rendered log-difference image D_hat directly corrupts the supervision signal, please report experiments with coarser pose streams or otherwise quantify sensitivity to this core input assumption.
minor comments (5)
  1. [Eq. (4)] The notation E(a,b) := ∫_a^b log(I'(t)) dt is unusual: the accumulated event image D approximates the integral of the time derivative of log intensity, i.e., log I(b) − log I(a). Please clarify the notation, for instance by writing d/dt log I(t) in the integrand.
  2. [Sec. 4.5] The sentence 'To fully utilize real-world event camera data, undistortion operation is performed as the accumulated image D is computed' is vague. Please state explicitly whether undistortion is applied to the event coordinates, to the accumulated image D, or to the rendered images, and where in the pipeline it occurs.
  3. [Sec. 5.1.2 and Table 2] TUM-VIE sequences are listed as part of the real-world evaluation and appear in qualitative figures (Fig. 4 and Fig. 9), but Table 2 reports only the five EDS scenes. Please add quantitative results for the TUM-VIE sequences or state explicitly why they are excluded from the numeric comparison.
  4. [Appendix B.2, Table 4] The comparison with EV-GS reports only mean PSNR and SSIM over four scenes. Please provide per-scene values, standard deviations, and a description of the evaluation protocol (e.g., whether the same color transform and test views were used for both methods), so that the comparison is reproducible and statistically meaningful.
  5. [Sec. 5.3] The metrics paragraph contains a typo ('It's worth noting') and a somewhat redundant explanation of why perceptual metrics are used. Consider tightening this paragraph and moving the linear color transform description to the experimental setup section, where it belongs.

Circularity Check

1 steps flagged · score 3.0 of 10

Mild disclosed circularity: quality metrics are computed after a linear color transform fitted to reference views; the core event-supervised 3DGS training is independent.

  1. fitted input called prediction [Sec. 5.3 (Metrics) and Sec. 5.6 (Limitations)]
    "the model cannot directly estimate absolute intensity images and requires a linear transformation with evaluation data as a reference. This linear transformation is required only during inference and does not impact the training."

    The evaluation protocol fits a linear color transformation to the evaluation/reference views before computing PSNR, SSIM, and LPIPS. Thus the final images compared in Tables 1 and 2 are not purely event-derived predictions; the unknown offset/scale is estimated from the ground-truth reference at test time. The central 3DGS scene optimization is unaffected, so this is a mild, disclosed calibration rather than a full reduction of the method to its inputs, but the headline quality numbers are partly fitted to the reference data by construction.

full rationale

The derivation chain is largely self-contained. The 3D Gaussian scene is optimized by comparing the rendered log-difference image D-hat (Eq. 6) with the accumulated event image D (Eq. 5) under an L1+SSIM loss (Eq. 9); the event accumulation, remosaicing, and spline pose interpolation are all defined from event data and poses, not from the target renderings. The event-to-video guided initialization uses a pretrained external model (E2VID) plus SfM; this is an external prior, not a result of this paper, and the ablation shows it improves rather than determines the outcome. Self-citations to [48] and [71] are contextual related-work mentions and do not carry the argument. The only step that is circular in the strict sense is the inference-time linear color transformation (Sec. 5.3, 5.6): because the model cannot estimate absolute intensity from relative log changes, a linear transform is fit using 'evaluation data as a reference' before metrics are computed. This means Tables 1 and 2 partially evaluate predictions that have been aligned to the reference views by construction. However, this is disclosed, standard for event-based view synthesis, limited to a global color/offset fit, and does not feed back into training or into the geometric content; the central claim of real-time 3DGS rendering from events is not reduced to a fit. The omission of concurrent event-based 3DGS baselines from the main tables is an experimental/validity concern, not a circularity.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The paper introduces no new physical entities or latent scene objects. Its central claim rests on a set of empirical and modeling assumptions: the event-accumulation to log-difference equivalence, static scenes, known/interpolable poses, a useful event-to-video prior, and a hand-set list of hyperparameters. The most significant hidden calibration is the inference-time linear color transform fitted to reference views, which is counted as a free parameter because it affects the reported metrics.

free parameters (5)
  • Loss weight lambda for SSIM term = Not stated in the paper
    Eq. (9) defines L = (1-lambda)*L1 + lambda*L_SSIM, but the selected lambda value is not reported. It controls the balance between reconstruction and perceptual losses and likely inherits a 3DGS default, but the paper does not confirm this.
  • Event contrast threshold delta (C+1/C-1) = C-1 = 0.25; symmetric C+1/C-1 = 1.0 for synthetic and EDS; asymmetric ratio 1.458 for TUM-VIE
    This scalar scales the accumulated event image D in Eq. (5). It is set per dataset based on sensor properties or prior work, and Appendix A.2.1 says the thresholds were 'co-optimized and trained' with the model, so it functions as a fitted input rather than a derived quantity.
  • Maximum event accumulation window fraction = 1% to 10% of total events
    Sec. 4.1 randomly samples start and end event indices with a maximum length ranging from 1% to 10% of the total event count. This hand-chosen range controls the mixture of global and local scene information available to the optimizer.
  • Opacity learning rate = 0.01 instead of the original 3DGS value of 0.05
    Appendix A.2 states that the only opacity learning rate change was from 0.05 to 0.01 to make training more stable under multi-view supervision with different accumulation lengths.
  • Inference-time linear color transform (scale and offset) = Per-scene, fitted to reference views
    Sec. 5.3 and Sec. 5.6 state that a linear transformation is applied to predictions to align them with reference data, and that this transformation is required only during inference. The coefficients are fit using evaluation reference frames, so part of the reported fidelity is calibrated against the test data themselves.
assumptions (6)
  • domain assumption The accumulated event image D over a random sub-trajectory is a valid approximation of the time integral of log-intensity change under the contrast threshold model.
    Eqs. (4)-(5) equate event accumulation with the integral of log-intensity change, ignoring threshold quantization, noise, and incomplete event coverage. This equivalence is the core supervision signal of the method.
  • domain assumption The scene is static and the camera poses can be recovered or interpolated from constant-rate pose samples.
    Sec. 5.1 requires high-frequency constant-rate poses, and Sec. 5.6 explicitly limits the method to static scenes. Dynamic object motion would break the event accumulation and reconstruction assumptions.
  • domain assumption The pretrained event-to-video model produces images whose texture and background content are sufficient for SfM to initialize Gaussian positions.
    Sec. 4.3 states that event-to-video generated images contain noise but retain texture and background information. The ablation in Tab. 3 supports the benefit empirically, but the assumption itself is not proven.
  • domain assumption Remosaicing with fixed 2x2 Bayer masks followed by logarithms preserves the relationship between rendered RGB images and color event accumulations.
    Sec. 4.2 uses fixed RGB masks and addition across channels to remosaic rendered images. This modeling choice is needed to compare RGB renderings with single-channel-per-pixel color event data.
  • standard math 3D Gaussian Splatting provides a differentiable rasterizer whose projected 2D covariance and alpha compositing correctly model image formation.
    The paper relies on the original 3DGS formulation [26] for covariance projection, sorting, and alpha blending without rederiving or verifying it.
  • domain assumption Cubic spline and spherical cubic spline interpolation accurately represent the real camera trajectory between sampled poses.
    Sec. 4.4 asserts that the interpolation maintains non-linear continuity in velocity and acceleration and closely approximates real-world camera motion, but no error analysis is provided for the pose interpolation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of EventSplat: 3D Gaussian Splatting from Moving Event Cameras for Real-time Rendering." pith.science (2026). https://pith.science/paper/WPQQERMI

@misc{pith2026241207293,
  author       = {Pith},
  title        = {Pith review of: EventSplat: 3D Gaussian Splatting from Moving Event Cameras for Real-time Rendering},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WPQQERMI}},
  note         = {Machine review of arXiv:2412.07293}
}
read the original abstract

We introduce a method for using event camera data in novel view synthesis via Gaussian Splatting. Event cameras offer exceptional temporal resolution and a high dynamic range. Leveraging these capabilities allows us to effectively address the novel view synthesis challenge in the presence of fast camera motion. For initialization of the optimization process, our approach uses prior knowledge encoded in an event-to-video model. We also use spline interpolation for obtaining high quality poses along the event camera trajectory. This enhances the reconstruction quality from fast-moving cameras while overcoming the computational limitations traditionally associated with event-based Neural Radiance Field (NeRF) methods. Our experimental evaluation demonstrates that our results achieve higher visual fidelity and better performance than existing event-based NeRF approaches while being an order of magnitude faster to render.

Figures

Figures reproduced from arXiv: 2412.07293 by the authors.

Figure 1
Figure 1. EventSplat derives 3D representations of scenes in the form of 3D Gaussians from event data, enabling fast real-time renderings [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overview of our 3D Gaussian Splatting training with moving event camera data. Event data streams from [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Generated Synthetic images comparing our work, event-based NeRF, and E2VID+3DGS qualitatively, with rendering times [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (3 more)
Figure 5
Figure 5. Figure 5: Qualitative comparisons of the images generated by [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Generated images are shown, qualitatively comparing our work, event-based NeRF, and E2VID+3DGS in all synthetic scenes. [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]
Figure 8
Figure 8. Figure 8: For each scene in the EDS dataset, we show generated images from two viewpoints alongside the ground truth image, comparing [PITH_FULL_IMAGE:figures/full_fig_p015_8.png]

Discussion (0). Continue with ORCID 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. GS2E: Gaussian Splatting is an Effective Data Generator for Event Stream Generation

    cs.CV 2025-05 conditional novelty 6.0 of 10

    A pipeline that turns sparse multi-view RGB images into a claimed 1,150-scene synthetic event dataset using 3D Gaussian Splatting rendering plus a stochastic event simulator.

  2. DeblurSplat: SfM-free 3D Gaussian Splatting with Event Camera for Robust Deblurring

    cs.CV 2025-09 conditional novelty 5.0 of 10

    A pose-free deblurring 3D Gaussian Splatting pipeline using DUSt3R point clouds, confidence-balanced sampling, and event-decoded latent image supervision.

Reference graph

Works this paper leans on

85 extracted references · 66 canonical work pages · cited by 2 Pith papers

  1. [1]

    Barron, Ben Mildenhall, Matthew Tancik, Pe- ter Hedman, Ricardo Martin-Brualla, and Pratul P

    Jonathan T. Barron, Ben Mildenhall, Matthew Tancik, Pe- ter Hedman, Ricardo Martin-Brualla, and Pratul P. Srini- vasan. Mip-NeRF: A multiscale representation for anti- aliasing neural radiance fields. ICCV, 2021. 1, 2

    2021

  2. [2]

    Barron, Ben Mildenhall, Dor Verbin, Pratul P

    Jonathan T. Barron, Ben Mildenhall, Dor Verbin, Pratul P. Srinivasan, and Peter Hedman. Mip-NeRF 360: Unbounded anti-aliased neural radiance fields. CVPR, 2022. 2

    2022

  3. [3]

    Barron, Ben Mildenhall, Dor Verbin, Pratul P

    Jonathan T. Barron, Ben Mildenhall, Dor Verbin, Pratul P. Srinivasan, and Peter Hedman. Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fields. In ICCV, 2023. 1

    2023

  4. [4]

    Event-based visual flow

    Ryad Benosman, Charles Clercq, Xavier Lagorce, Sio-Hoi Ieng, and Chiara Bartolozzi. Event-based visual flow. TNNLS, 2013. 2

    2013

  5. [5]

    Accurate detection of demosaic- ing regularity for digital image forensics

    Hong Cao and Alex C Kot. Accurate detection of demosaic- ing regularity for digital image forensics. IEEE Trans. Inf. Forensics Secur, 2009. 4

    2009

  6. [6]

    Tensorf: Tensorial radiance fields

    Anpei Chen, Zexiang Xu, Andreas Geiger, Jingyi Yu, and Hao Su. Tensorf: Tensorial radiance fields. In ECCV. Springer, 2022. 2

    work page 2022

  7. [7]

    A survey on 3d gaussian splatting

    Guikun Chen and Wenguan Wang. A survey on 3d gaussian splatting. arXiv preprint arXiv:2401.03890, 2024. 2, 3

    arXiv 2024

  8. [8]

    Revisiting Event-Based Video Frame Interpolation

    Jiaben Chen, Yichen Zhu, Dongze Lian, Jiaqi Yang, Yifu Wang, Renrui Zhang, Xinhang Liu, Shenhan Qian, Laurent Kneip, and Shenghua Gao. Revisiting Event-Based Video Frame Interpolation. In IROS, 2023. 1

    work page 2023

Show all 85 references
  1. [9]

    MobileNeRF: Exploiting the polygon ras- terization pipeline for efficient neural field rendering on mo- bile architectures

    Zhiqin Chen, Thomas Funkhouser, Peter Hedman, and An- drea Tagliasacchi. MobileNeRF: Exploiting the polygon ras- terization pipeline for efficient neural field rendering on mo- bile architectures. In arXiv, 2022. 1, 2

    2022

  2. [10]

    Text-to-3D us- ing Gaussian Splatting

    Zilong Chen, Feng Wang, and Huaping Liu. Text-to-3D us- ing Gaussian Splatting. arXiv preprint arXiv:2309.16585 ,

    arXiv

  3. [11]

    E2gs: Event enhanced gaussian splatting

    Hiroyuki Deguchi, Mana Masuda, Takuya Nakabayashi, and Hideo Saito. E2gs: Event enhanced gaussian splatting. In 2024 IEEE International Conference on Image Processing (ICIP), 2024. 2

    2024

  4. [12]

    Depth-supervised NeRF: Fewer views and faster training for free

    Kangle Deng, Andrew Liu, Jun-Yan Zhu, and Deva Ra- manan. Depth-supervised NeRF: Fewer views and faster training for free. In CVPR, 2022. 2

    2022

  5. [13]

    Dy- namic obstacle avoidance for quadrotors with event cameras

    Davide Falanga, Kevin Kleber, and Davide Scaramuzza. Dy- namic obstacle avoidance for quadrotors with event cameras. Sci. Robot, 2020. 1

    2020

  6. [14]

    Event- based, 6-dof camera tracking from photometric depth maps

    Guillermo Gallego, Jon EA Lund, Elias Mueggler, Henri Rebecq, Tobi Delbruck, and Davide Scaramuzza. Event- based, 6-dof camera tracking from photometric depth maps. TPAMI, 2017. 2

    2017

  7. [15]

    A unifying contrast maximization framework for event cam- eras, with applications to motion, depth, and optical flow es- timation

    Guillermo Gallego, Henri Rebecq, and Davide Scaramuzza. A unifying contrast maximization framework for event cam- eras, with applications to motion, depth, and optical flow es- timation. In CVPR, 2018. 1

    2018

  8. [16]

    Davison, Jörg Conradt, Kostas Daniilidis, and Da- vide Scaramuzza

    Guillermo Gallego, Tobi Delbrück, Garrick Orchard, Chiara Bartolozzi, Brian Taba, Andrea Censi, Stefan Leutenegger, Andrew J. Davison, Jörg Conradt, Kostas Daniilidis, and Da- vide Scaramuzza. Event-based vision: A survey. PAMI,

  9. [17]

    Garbin, Marek Kowalski, Matthew Johnson, Jamie Shotton, and Julien Valentin

    Stephan J. Garbin, Marek Kowalski, Matthew Johnson, Jamie Shotton, and Julien Valentin. FastNeRF: High-Fidelity Neural Rendering at 200FPS. In ICCV, 2021. 2

    2021

  10. [18]

    Asynchronous, photometric feature track- ing using events and frames

    Daniel Gehrig, Henri Rebecq, Guillermo Gallego, and Da- vide Scaramuzza. Asynchronous, photometric feature track- ing using events and frames. In ECCV, 2018. 2

    2018

  11. [19]

    End-to-end learning of rep- resentations for asynchronous event-based data

    Daniel Gehrig, Antonio Loquercio, Konstantinos G Derpa- nis, and Davide Scaramuzza. End-to-end learning of rep- resentations for asynchronous event-based data. In ICCV, 2019

    2019

  12. [20]

    Eklt: Asynchronous photometric feature tracking using events and frames

    Daniel Gehrig, Henri Rebecq, Guillermo Gallego, and Da- vide Scaramuzza. Eklt: Asynchronous photometric feature tracking using events and frames. IJCV, 2020

    2020

  13. [21]

    Recurrent vision transformers for object detection with event cameras

    Mathias Gehrig and Davide Scaramuzza. Recurrent vision transformers for object detection with event cameras. In CVPR, 2023. 2

    2023

  14. [22]

    E-raft: Dense optical flow from event cam- eras

    Mathias Gehrig, Mario Millhäusler, Daniel Gehrig, and Da- vide Scaramuzza. E-raft: Dense optical flow from event cam- eras. In 3DV, 2021. 2

    2021

  15. [23]

    Srinivasan, Ben Mildenhall, Jonathan T

    Peter Hedman, Pratul P. Srinivasan, Ben Mildenhall, Jonathan T. Barron, and Paul Debevec. Baking neural ra- diance fields for real-time view synthesis. In ICCV, 2021. 2

    2021

  16. [24]

    Event-aided direct sparse odometry

    Javier Hidalgo-Carrió, Guillermo Gallego, and Davide Scaramuzza. Event-aided direct sparse odometry. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022. 5

    2022

  17. [25]

    Ev-NeRF: Event based neural radiance field

    Inwoo Hwang, Junho Kim, and Young Min Kim. Ev-NeRF: Event based neural radiance field. In WACV, 2023. 1, 2

    2023

  18. [26]

    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. ToG, 2023. 1, 2, 3, 12

    2023

  19. [27]

    Real-time 3d reconstruction and 6-dof tracking with an event camera

    Hanme Kim, Stefan Leutenegger, and Andrew J Davison. Real-time 3d reconstruction and 6-dof tracking with an event camera. In CVPR, 2016. 1

    2016

  20. [28]

    N-imagenet: Towards robust, fine-grained object recognition with event cameras

    Junho Kim, Jaehyeok Bae, Gangin Park, Dongsu Zhang, and Young Min Kim. N-imagenet: Towards robust, fine-grained object recognition with event cameras. In ICCV, 2021. 2

    2021

  21. [29]

    Demosaicing: Image reconstruction from color CCD samples

    Ron Kimmel. Demosaicing: Image reconstruction from color CCD samples. IEEE Trans. Image Process, 1999. 4

    1999

  22. [30]

    TUM-VIE: The TUM Stereo Visual-Inertial Event Dataset

    Simon Klenk, Jason Chui, Nikolaus Demmel, and Daniel Cremers. TUM-VIE: The TUM Stereo Visual-Inertial Event Dataset. In IROS, 2021. 5

    2021

  23. [31]

    E-nerf: Neural radiance fields from a mov- ing event camera

    Simon Klenk, Lukas Koestler, Davide Scaramuzza, and Daniel Cremers. E-nerf: Neural radiance fields from a mov- ing event camera. RA-L, 2023. 1, 2

    2023

  24. [32]

    AdaNeRF: Adaptive sam- pling for real-time rendering of neural radiance fields

    Andreas Kurz, Thomas Neff, Zhaoyang Lv, Michael Zoll- höfer, and Markus Steinberger. AdaNeRF: Adaptive sam- pling for real-time rendering of neural radiance fields. In ECCV, 2022. 2

    2022

  25. [33]

    Pulsar: Efficient Sphere-Based Neural Rendering

    Christoph Lassner and Michael Zollhofer. Pulsar: Efficient Sphere-Based Neural Rendering. In CVPR, 2021. 1 9

    2021

  26. [34]

    Asynchronous spatio-temporal memory net- work for continuous event-based object detection.TIP, 2022

    Jianing Li, Jia Li, Lin Zhu, Xijie Xiang, Tiejun Huang, and Yonghong Tian. Asynchronous spatio-temporal memory net- work for continuous event-based object detection.TIP, 2022. 2

    2022

  27. [35]

    Demosaicing by successive approximation

    Xin Li. Demosaicing by successive approximation. IEEE Trans. Image Process, 2005. 4

    2005

  28. [36]

    Image demosaic- ing: A systematic survey

    Xin Li, Bahadir Gunturk, and Lei Zhang. Image demosaic- ing: A systematic survey. In VCIP. SPIE, 2008. 4

    2008

  29. [37]

    A 128× 128 120 db 15 µs latency asynchronous temporal con- trast vision sensor

    Patrick Lichtsteiner, Christoph Posch, and Tobi Delbruck. A 128× 128 120 db 15 µs latency asynchronous temporal con- trast vision sensor. JSSC, 2008. 1

    2008

  30. [38]

    Learning Event-Driven Video Deblurring and Interpolation

    Songnan Lin, Jiawei Zhang, Jinshan Pan, Zhe Jiang, Dongqing Zou, Yongtian Wang, Jing Chen, and Jimmy Ren. Learning Event-Driven Video Deblurring and Interpolation. In ECCV, 2020. 1

    2020

  31. [39]

    BACON: Band-limited coordinate networks for multiscale scene representation

    David B Lindell, Dave Van Veen, Jeong Joon Park, and Gor- don Wetzstein. BACON: Band-limited coordinate networks for multiscale scene representation. In CVPR, 2022. 2

    2022

  32. [40]

    Align your gaussians: Text-to-4d with dynamic 3d gaussians and composed diffusion models

    Huan Ling, Seung Wook Kim, Antonio Torralba, Sanja Fi- dler, and Karsten Kreis. Align your gaussians: Text-to-4d with dynamic 3d gaussians and composed diffusion models. arXiv preprint arXiv:2312.13763, 2023. 2

    2023 arXiv

  33. [41]

    Perceptual assessment of demosaicing algorithm performance

    Philippe Longere, Xuemei Zhang, Peter B Delahunt, and David H Brainard. Perceptual assessment of demosaicing algorithm performance. Proceedings of the IEEE, 2002. 4

    2002

  34. [42]

    Robust e-NeRF: NeRF from Sparse & Noisy Events under Non-Uniform Motion,

    Weng Fei Low and Gim Hee Lee. Robust e-NeRF: NeRF from Sparse & Noisy Events under Non-Uniform Motion,

  35. [43]

    Li Ma, Xiaoyu Li, Jing Liao, Qi Zhang, Xuan Wang, Jue Wang, and Pedro V . Sander. Deblur-NeRF: Neural Radiance Fields From Blurry Images. In CVPR, 2022. 1

    2022

  36. [44]

    High- quality linear interpolation for demosaicing of bayer- patterned color images

    Henrique S Malvar, Li-wei He, and Ross Cutler. High- quality linear interpolation for demosaicing of bayer- patterned color images. In ICASSP, 2004. 4

    2004

  37. [45]

    Gaussian splatting slam

    Hidenobu Matsuki, Riku Murai, Paul HJ Kelly, and An- drew J Davison. Gaussian splatting slam. arXiv preprint arXiv:2312.06741, 2023. 2

    2023 arXiv

  38. [46]

    Srinivasan, Matthew Tancik, Jonathan T

    Ben Mildenhall, Pratul P. Srinivasan, Matthew Tancik, Jonathan T. Barron, Ravi Ramamoorthi, and Ren Ng. NeRF: representing scenes as neural radiance fields for view synthe- sis. ECCV, 2021. 1, 2

    2021

  39. [47]

    Srinivasan, and Jonathan T

    Ben Mildenhall, Peter Hedman, Ricardo Martin-Brualla, Pratul P. Srinivasan, and Jonathan T. Barron. NeRF in the Dark: High Dynamic Range View Synthesis From Noisy Raw Images. In CVPR, 2022. 1

    2022

  40. [48]

    Derpanis, Jonathan Kelly, Marcus A

    Ashkan Mirzaei, Tristan Aumentado-Armstrong, Konstanti- nos G. Derpanis, Jonathan Kelly, Marcus A. Brubaker, Igor Gilitschenski, and Alex Levinshtein. SPIn-NeRF: Multiview Segmentation and Perceptual Inpainting With Neural Radi- ance Fields. In CVPR, 2023

    2023

  41. [49]

    Instant neural graphics primitives with a multires- olution hash encoding

    Thomas Müller, Alex Evans, Christoph Schied, and Alexan- der Keller. Instant neural graphics primitives with a multires- olution hash encoding. TOG, 2022. 1, 2

    2022

  42. [50]

    A spiking neural network model of 3d perception for event-based neuromorphic stereo vision systems

    Marc Osswald, Sio-Hoi Ieng, Ryad Benosman, and Giacomo Indiveri. A spiking neural network model of 3d perception for event-based neuromorphic stereo vision systems. Sci. Rep., 2017. 1

    2017

  43. [51]

    Bringing a blurry frame alive at high frame-rate with an event camera

    Liyuan Pan, Cedric Scheerlinck, Xin Yu, Richard Hartley, Miaomiao Liu, and Yuchao Dai. Bringing a blurry frame alive at high frame-rate with an event camera. In CVPR,

  44. [52]

    Learning to detect objects with a 1 megapixel event camera

    Etienne Perot, Pierre De Tournemire, Davide Nitti, Jonathan Masci, and Amos Sironi. Learning to detect objects with a 1 megapixel event camera. NeurIPS, 2020. 2

    2020

  45. [53]

    Barron, and Ben Milden- hall

    Ben Poole, Ajay Jain, Jonathan T. Barron, and Ben Milden- hall. DreamFusion: Text-to-3D using 2D Diffusion, 2022. 2, 3

    2022

  46. [54]

    Evo: A geometric approach to event- based 6-dof parallel tracking and mapping in real time.RA-L,

    Henri Rebecq, Timo Horstschäfer, Guillermo Gallego, and Davide Scaramuzza. Evo: A geometric approach to event- based 6-dof parallel tracking and mapping in real time.RA-L,

  47. [55]

    ESIM: an open event camera simulator

    Henri Rebecq, Daniel Gehrig, and Davide Scaramuzza. ESIM: an open event camera simulator. In CoRL. PMLR,

  48. [56]

    High speed and high dynamic range video with an event camera

    Henri Rebecq, René Ranftl, Vladlen Koltun, and Davide Scaramuzza. High speed and high dynamic range video with an event camera. PAMI, 2019. 5

    2019

  49. [57]

    KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPs

    Christian Reiser, Songyou Peng, Yiyi Liao, and Andreas Geiger. KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPs. In ICCV, 2021. 2

    2021

  50. [58]

    Srinivasan, Ben Mildenhall, Andreas Geiger, Jonathan T

    Christian Reiser, Richard Szeliski, Dor Verbin, Pratul P. Srinivasan, Ben Mildenhall, Andreas Geiger, Jonathan T. Barron, and Peter Hedman. MERF: Memory-efficient radi- ance fields for real-time view synthesis in unbounded scenes. In arXiv, 2023. 2

    2023

  51. [59]

    EventNeRF: Neural radiance fields from a single colour event camera

    Viktor Rudnev, Mohamed Elgharib, Christian Theobalt, and Vladislav Golyanik. EventNeRF: Neural radiance fields from a single colour event camera. In CVPR, 2023. 1, 2, 6, 12

    2023

  52. [60]

    Plenoxels: Radiance fields without neural networks

    Sara Fridovich-Keil and Alex Yu, Matthew Tancik, Qinhong Chen, Benjamin Recht, and Angjoo Kanazawa. Plenoxels: Radiance fields without neural networks. In CVPR, 2022. 2

    2022

  53. [61]

    HATS: Histograms of aver- aged time surfaces for robust event-based object classifica- tion

    Amos Sironi, Manuele Brambilla, Nicolas Bourdis, Xavier Lagorce, and Ryad Benosman. HATS: Histograms of aver- aged time surfaces for robust event-based object classifica- tion. In CVPR, 2018. 2

    2018

  54. [62]

    Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapes

    Towaki Takikawa, Joey Litalien, Kangxue Yin, Karsten Kreis, Charles Loop, Derek Nowrouzezahrai, Alec Jacobson, Morgan McGuire, and Sanja Fidler. Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapes. In CVPR, 2021. 2

    2021

  55. [63]

    Time lens: Event-based video frame interpo- lation

    Stepan Tulyakov, Daniel Gehrig, Stamatios Georgoulis, Julius Erbach, Mathias Gehrig, Yuanyou Li, and Davide Scaramuzza. Time lens: Event-based video frame interpo- lation. In CVPR, 2021. 1

    2021

  56. [64]

    Time Lens++: Event-Based Frame Interpolation With Para- metric Non-Linear Flow and Multi-Scale Fusion

    Stepan Tulyakov, Alfredo Bochicchio, Daniel Gehrig, Sta- matios Georgoulis, Yuanyou Li, and Davide Scaramuzza. Time Lens++: Event-Based Frame Interpolation With Para- metric Non-Linear Flow and Multi-Scale Fusion. In CVPR,

  57. [65]

    Barron, and Pratul P

    Dor Verbin, Peter Hedman, Ben Mildenhall, Todd Zickler, Jonathan T. Barron, and Pratul P. Srinivasan. Ref-NeRF: Structured view-dependent appearance for neural radiance fields. CVPR, 2022. 2 10

    2022

  58. [66]

    Evggs: A collaborative learning framework for event-based generalizable gaussian splatting

    Jiaxu Wang, Junhao He, Ziyi Zhang, Mingyuan Sun, Jingkai Sun, and Renjing Xu. Evggs: A collaborative learning framework for event-based generalizable gaussian splatting. arXiv preprint arXiv:2405.14959, 2024. 2

    2024 arXiv

  59. [67]

    Wang, E.P

    Z. Wang, E.P. Simoncelli, and A.C. Bovik. Multiscale Struc- tural Similarity for Image Quality Assessment. In ACSSC,

  60. [68]

    Eadeblur-gs: Event assisted 3d deblur reconstruction with gaussian splatting

    Yuchen Weng, Zhengwen Shen, Ruofan Chen, Qi Wang, and Jun Wang. Eadeblur-gs: Event assisted 3d deblur reconstruction with gaussian splatting. arXiv preprint arXiv:2407.13520, 2024. 2

    2024 arXiv

  61. [69]

    4D Gaussian Splatting for Real-Time Dynamic Scene Ren- dering

    Guanjun Wu, Taoran Yi, Jiemin Fang, Lingxi Xie, Xiaopeng Zhang, Wei Wei, Wenyu Liu, Qi Tian, and Xinggang Wang. 4D Gaussian Splatting for Real-Time Dynamic Scene Ren- dering. arXiv preprint arXiv:2310.08528, 2023. 1

    2023 arXiv

  62. [70]

    Ev-gs: Event-based gaussian splatting for effi- cient and accurate radiance field rendering

    Jingqian Wu, Shuo Zhu, Chutian Wang, and Edmund Y Lam. Ev-gs: Event-based gaussian splatting for effi- cient and accurate radiance field rendering. arXiv preprint arXiv:2407.11343, 2024. 2, 12, 14

    2024 arXiv

  63. [71]

    Leod: Label-efficient object detection for event cameras

    Ziyi Wu, Mathias Gehrig, Qing Lyu, Xudong Liu, and Igor Gilitschenski. Leod: Label-efficient object detection for event cameras. arXiv preprint arXiv:2311.17286, 2023. 2

    2023 arXiv

  64. [72]

    Event3dgs: Event-based 3d gaussian splatting for high-speed robot egomotion

    Tianyi Xiong, Jiayi Wu, Botao He, Cornelia Fermuller, Yiannis Aloimonos, Heng Huang, and Christopher Metzler. Event3dgs: Event-based 3d gaussian splatting for high-speed robot egomotion. In 8th Annual Conference on Robot Learn- ing, 2024. 2

    2024

  65. [73]

    Real- time Photorealistic Dynamic Scene Representation and Ren- dering with 4D Gaussian Splatting

    Zeyu Yang, Hongye Yang, Zijie Pan, and Li Zhang. Real- time Photorealistic Dynamic Scene Representation and Ren- dering with 4D Gaussian Splatting. In ICLR, 2024. 2

    2024

  66. [74]

    BakedSDF: Meshing Neural SDFs for Real-Time View Synthesis

    Lior Yariv, Peter Hedman, Christian Reiser, Dor Verbin, Pratul P Srinivasan, Richard Szeliski, Jonathan T Barron, and Ben Mildenhall. BakedSDF: Meshing Neural SDFs for Real-Time View Synthesis. In SIGGRAPH, 2023. 1

    2023

  67. [75]

    Gaussiandreamer: Fast generation from text to 3d gaussians by bridging 2d and 3d diffusion models

    Taoran Yi, Jiemin Fang, Junjie Wang, Guanjun Wu, Lingxi Xie, Xiaopeng Zhang, Wenyu Liu, Qi Tian, and Xinggang Wang. Gaussiandreamer: Fast generation from text to 3d gaussians by bridging 2d and 3d diffusion models. In CVPR,

  68. [76]

    PlenOctrees for real-time rendering of neural radiance fields

    Alex Yu, Ruilong Li, Matthew Tancik, Hao Li, Ren Ng, and Angjoo Kanazawa. PlenOctrees for real-time rendering of neural radiance fields. In ICCV, 2021. 2

    2021

  69. [77]

    Evagaussians: Event stream assisted gaussian splatting from blurry images

    Wangbo Yu, Chaoran Feng, Jiye Tang, Xu Jia, Li Yuan, and Yonghong Tian. Evagaussians: Event stream assisted gaussian splatting from blurry images. arXiv preprint arXiv:2405.20224, 2024. 2

    2024 arXiv

  70. [78]

    Zhiyang Yu, Yu Zhang, Deyuan Liu, Dongqing Zou, Xijun Chen, Yebin Liu, and Jimmy S. Ren. Training weakly super- vised video frame interpolation with events. In ICCV, 2021. 1

    2021

  71. [79]

    Efros, Eli Shecht- man, and Oliver Wang

    Richard Zhang, Phillip Isola, Alexei A. Efros, Eli Shecht- man, and Oliver Wang. The Unreasonable Effectiveness of Deep Features as a Perceptual Metric. In CVPR, 2018. 5

    2018

  72. [80]

    Elite-evgs: Learning event-based 3d gaussian splatting by distilling event-to-video priors

    Zixin Zhang, Kanghao Chen, and Lin Wang. Elite-evgs: Learning event-based 3d gaussian splatting by distilling event-to-video priors. arXiv preprint arXiv:2409.13392 ,

    arXiv

  73. [81]

    Semi-Dense 3D Re- construction with a Stereo Event Camera

    Yi Zhou, Guillermo Gallego, Henri Rebecq, Laurent Kneip, Hongdong Li, and Davide Scaramuzza. Semi-Dense 3D Re- construction with a Stereo Event Camera. In ECCV, 2018. 2

    2018

  74. [82]

    Event-based stereo visual odometry

    Yi Zhou, Guillermo Gallego, and Shaojie Shen. Event-based stereo visual odometry. T-RO, 2021. 2

    2021

  75. [83]

    Ev-flownet: Self- supervised optical flow estimation for event-based cameras

    Alex Zihao Zhu and Liangzhe Yuan. Ev-flownet: Self- supervised optical flow estimation for event-based cameras. In RSS, 2018. 2

    2018

  76. [84]

    AutoReCon: Neural Architecture Search- based Reconstruction for Data-free Compression, 2021

    Baozhou Zhu, Peter Hofstee, Johan Peltenburg, Jinho Lee, and Zaid Alars. AutoReCon: Neural Architecture Search- based Reconstruction for Data-free Compression, 2021. 1

    2021

  77. [85]

    Zwicker, H

    M. Zwicker, H. Pfister, J. van Baar, and M. Gross. EW A volume splatting. In VIS, 2001. 2, 3 11 A. Implementation Details A.1. Algorithm Our optimization and densification algorithm is shown in Algorithm 1. All modifications compared to the original Gaussian Splatting process ...

    2001

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.