Pith. sign in

REVIEW 3 major objections 5 minor 59 references

FoveaSPAD: Exploiting Depth Priors for Adaptive and Efficient Single-Photon 3D Imaging

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read FoveaSPAD claims that guiding SPAD histogram capture with depth priors cuts raw data by 1548-fold while keeping depth accuracy and improving ambient-light resilience.

desk verdict A genuinely useful idea about adaptive SPAD histogram capture that deserves review, but the theory has correctable algebra errors and the headline efficiency numbers are conditional on an unquantified prior-accuracy assumption. read the letter →

arxiv 2412.02052 v1 pith:F7JNSRZR submitted 2024-12-03 eess.IV cs.CV

classification eess.IVcs.CV
keywords single-photonavalanchediodetime-of-flightLiDARfoveationdepthpriorphotontiminghistogrammemoryefficiencyambientlightrejectioncomputationalimaging
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

FoveaSPAD argues that a SPAD time-of-flight camera does not need to store the full photon-timing histogram. If an external depth prior, from a monocular network, optical flow, or a coarse scan, places the laser echo inside a small window of M histogram bins, the sensor can capture only those bins. The paper derives why this works: memory foveation keeps the same per-bin SNR while cutting stored data by M/N, and it raises the signal-to-background ratio because ambient photons arriving before the signal bin no longer contribute to detector dead time. Depth foveation instead packs a fixed memory budget into the window to improve depth resolution, at an SNR cost that more laser cycles can repay. If the claim holds, SPAD LiDAR can approach the depth quality of full histograms with a fraction of the bandwidth, which matters for autonomous vehicles and power-constrained sensors.

What carries the argument

The central object is the foveation window: a per-pixel subset of M histogram bins (M much smaller than N) placed around an estimated depth from a prior. The argument runs through two identities. First, SNR scales with bin width, so keeping the original bin width inside the M-bin window preserves SNR while storing only M/N of the histogram. Second, SBR depends on the probability that an ambient photon in an earlier bin resets the detector before the laser echo; Eq. (6) shows foveation removes those early-bin terms from the denominator, and perfect foveation reduces SBR to the direct signal-vs-background ratio. Depth foveation reuses the memory savings to place more, narrower bins inside the window, which is how the paper converts bandwidth savings into resolution.

What would settle it

Use a static scene with known ground-truth depth, introduce a controlled bias into the depth prior that moves the window more than half its width away from the true peak, and compare foveated depth error to full-histogram depth error. If the foveated capture still recovers the correct peak, or if the error grows only as fast as bin width rather than jumping to the noise floor, the paper's dependence on prior accuracy would be contradicted; the paper's own worst-case analysis predicts the jump.

Watch

Extended reading notes

Core claim

The paper's central claim is that foveated capture, gating each SPAD pixel to a window of M bins centered on a depth prior, converts the SPAD histogram bottleneck into a tunable trade. In memory foveation the bin width stays T/N and only the M bins around the predicted peak are stored; SNR is unchanged and memory falls by M/N. In depth foveation the window is re-divided into the same number of bins one would otherwise spread over the full range, so depth resolution improves while SNR drops by sqrt(M/N), recoverable by increasing the number of laser cycles. Under ambient light, memory foveation raises SBR because photons arriving before the foveation window no longer reset the detector; with a perfect prior the denominator's prior-bin dependence vanishes, leaving SBR proportional to the direct signal-vs-background odds. The paper supports this with simulations on an indoor RGB-D benchmark using a monocular prior, spatio-temporal quantized sampling that yields a 1548-fold memory reduction, optical-flow-driven foveation on driving scenes, and hardware emulation on real SPAD datasets where even simple peak detection improves after foveation.

Load-bearing premise

Everything rests on the depth prior being accurate enough that the true laser echo lands inside the M-bin foveation window: if the prior is biased by more than half the window width, the saved memory and the SNR/SBR gains collapse because the sensor never records the signal.

Editorial extensions

If this is right

  • On an indoor RGB-D benchmark with a monocular depth prior, memory foveation at 1/16 of the histogram bins keeps depth errors close to full-resolution SPAD simulation, and depth foveation with the same memory budget improves resolution over uniformly spread limited bins.
  • Spatio-temporal foveation that samples a few pixels per quantized depth bucket and foveates each in time achieves a 1548-fold memory reduction while still recovering scene depth.
  • Memory foveation extends the operable ambient-light range: hardware emulation shows the foveated photon cube has fewer background detections and a simple maxima estimator recovers structure that full-histogram maxima misses.
  • Optical-flow-driven foveation transfers the depth prior between frames for moving scenes, with a noise-floor comparison that resets pixels whose foveated window has drifted off the signal.
  • Superpixel-based foveation on real SPAD scans without a co-located camera reduces per-pixel memory by about 64 times for over 99 percent of pixels, using one full histogram per segment to anchor the window.

Reading between the lines

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

  • A natural extension the paper does not explore is making the window size adaptive to prior confidence: pixels with high prior uncertainty could keep larger M, trading memory for robustness exactly where the prior is weakest.
  • The same M/N storage saving implies that once per-pixel gating hardware exists, a SPAD array could raise spatial resolution or frame rate by roughly N/M without increasing off-chip bandwidth, for scenes with accurate priors.
  • Because the related-work section notes compressive histogramming and sketching are complementary, a plausible next test is combining foveated capture with a compressive projection to compound the bandwidth reduction; the paper does not run that experiment.
  • The paper's worst-case probability analysis suggests a quantitative failure boundary: scenes with multipath effects degrade catastrophically only when the single-bounce probability and multipath probability satisfy specific relations, so a calibration experiment sweeping prior bias against M could map acceptable operating conditions.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The manuscript introduces FoveaSPAD, a family of adaptive capture policies for SPAD-based LiDAR in which an external depth prior (monocular depth, optical flow, or low-resolution sampling) is used to restrict or reallocate histogram bins around an expected echo time. The authors distinguish memory foveation (fewer bins at full width) and depth foveation (fixed bin budget concentrated in a window), and they claim theoretical gains in SNR/SBR, memory, and depth resolution. The results are demonstrated on simulated datasets (NYUv2, CARLA) and hardware emulation of real SPAD data (Lindell et al., Gutierrez-Barragan et al.), including a reported 1548x memory savings in a spatio-temporal variant. The paper explicitly acknowledges that the strategies are dependent on the accuracy of the depth prior, and it provides a worst-case stochastic analysis in Sec. VIII.

Significance. Foveated capture is an important and timely idea for SPAD LiDAR because raw histogram bandwidth is a known bottleneck, and the paper makes a credible case that a prior can shift the sampling budget. The simulations and emulations are performed on standard public datasets, and the hardware-emulation results against real SPAD data add value. The reported memory savings are substantial. However, the stated theoretical support for the efficiency gains contains concrete errors in Eqs. (4) and (7), and the empirical claims are conditional on the prior correctly localizing the histogram peak, a condition that is acknowledged but never quantified. If these issues are corrected and an explicit error-propagation analysis is added, the paper could be a useful contribution to adaptive single-photon imaging.

major comments (3)
  1. [Sec. III-C and Sec. VIII] The central efficiency claim is conditional on the foveation window containing the true histogram peak. Eqs. (2)-(8) and the simulation protocol all assume this event, and Sec. III-A states that the strategies are fundamentally dependent on prior accuracy. The worst-case analysis in Sec. VIII, Eqs. (9)-(10), is a stochastic expression for total depth-detection failure under multipath and noise, but it is not tied to measured prior-error statistics and is not validated experimentally. The optical-flow results in Sec. VI already exhibit a failure of the conditioning event: in the first CARLA scene at M=1/10N the reported RMSE is 101.9 m. Without a quantitative characterization of window-inclusion probability as a function of prior error, the claim that foveation 'maintains depth accuracy' is not established for realistic biased priors. This is a load-bearing gap because all reported memory and SNR gains depend on the peak being captured.
  2. [Eq. (4)] Eq. (4) states that depth foveation requires C_new/C >= N^2/M^2 to match conventional SNR. From Eq. (3), SNR is proportional to C sqrt(M T / N^2), which equals C sqrt(M/N) sqrt(T/N), while the conventional SNR in Eq. (1) is C sqrt(T/N). Equating these gives C_new/C = sqrt(N/M), not (N/M)^2. The current expression overstates the required exposure increase by a factor of (N/M)^{3/2} for typical M << N. This error directly affects the theoretical claim in Sec. III-C that depth foveation can be compensated by more laser cycles.
  3. [Eq. (7)] Eq. (7) omits the denominator in the SBR expression. Starting from Eq. (6) with j=i, the numerator becomes (1 - e^{-(Phi_sig+Phi_bkg)}), but the denominator is not 1: it contains p^i_bkg = (1 - e^{-Phi_bkg}) e^{-Sum_{1}^{i-1} Phi_bkg}. The perfect-foveation limit therefore still depends on the ambient level through the probability that a background photon is detected at the correct bin. The text's assertion that foveation 'removes the dependence on prior photon arrival' is only valid in the limit Phi_bkg approaching 0, which is the opposite of the strong-ambient-light regime this subsection addresses. This error weakens the claimed SBR advantage of memory foveation.
minor comments (5)
  1. [Eq. (2)] The mathematical expression in Eq. (2) is garbled and should be rewritten for readability; the intended formula appears to be SNR proportional to sqrt(T/N).
  2. [Sec. III-C] The notation is overloaded: in the paragraph after Eq. (4), 'the foveated bins N are given to us' uses N to mean the earlier M, which is confusing.
  3. [Table II] The Table II header contains the typo 'ERRROR'; please correct to 'ERROR'.
  4. [Affiliation] The author affiliation line lists 'Gainsville, FL'; the correct spelling of the city is 'Gainesville'.
  5. [Eq. (5)] Eq. (5) contains an extra closing parenthesis after (1 - e^{-(Phi_sig+Phi_bkg)}), resulting in '(1 - e^{-(Phi_sig+Phi_bkg)}))'.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation found: the foveation gains are explicitly conditioned on prior accuracy, and the accuracy claims are validated on external public data.

full rationale

The derivation chain is not circular. Section III-C explicitly conditions the SNR/SBR model on the foveation window containing the histogram peak: "we will not make any assumption as to how the foveated bins M were obtained and instead just characterize the advantage of these, given that the desired histogram peak is captured by these bins." This is a transparent conditioning assumption, not a hidden equivalence. The memory-reduction factors are bin-count arithmetic (M/N and pixel sparsity), and the depth-accuracy claims are empirical results from simulations and hardware emulation on NYUv2, CARLA, Lindell et al., and Gutierrez-Barragan et al. public data, with ground truth produced from full-resolution SPAD histograms. The monocular-prior calibration in Section IV fits a scaling polynomial using a small set of full-resolution SPAD pixels, but the final reported depths are the SPAD histogram peaks, not the fitted prior; this is calibration, not a fitted quantity being renamed as a prediction. The author-co-authored references [46]-[48] supply the standard SPAD Poisson/binomial imaging model and simulation code; the foveation result does not reduce to those citations, and no uniqueness theorem or ansatz is imported from them. The paper candidly acknowledges the central conditioning in Sections III-A and VIII, and reports a failure of the conditioning event in the first CARLA optical-flow scene (RMSE 101.9 m at M = 1/10 N). Therefore no specific step can be exhibited in which an output is equivalent by definition or by self-citation to an input. The algebraic issues in Eqs. (4) and (7) and the untested worst-case stochastic model in Section VIII are correctness and completeness concerns, not circularity, and do not change this verdict.

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

The central foveation claim rests mainly on the Poisson single-photon model borrowed from prior work and on the assumption that an accurate depth prior places the histogram peak inside the foveated window. No new physical entities are introduced. Several user-chosen hyperparameters (window fraction, bin count, scaling fit) shape the reported efficiency numbers.

free parameters (5)
  • Foveation window fraction M/N = 1/16, 1/8, 1/4
    User-set fraction of histogram bins kept in each pixel; controls memory/accuracy trade-off in all experiments.
  • Depth foveation bin count N' = 16, 32, 64
    Number of bins packed into the foveated window; chosen by hand in Tables II and III.
  • Monocular depth scaling polynomial coefficients = Not stated numerically
    Polynomial fit maps ZoeDepth monocular output to SPAD metric depth using a subset of full-resolution pixels; coefficients are fit to data per scene or globally.
  • Optical-flow noise-floor threshold = Not stated
    Threshold used to detect foveated windows close to noise floor and trigger full-histogram recomputation; value not specified in text.
  • Quantization bucket count and samples per bucket = 64 buckets, 50 points
    Hyperparameters for spatio-temporal sampling that set the sparsity level.
assumptions (6)
  • domain assumption Photon arrivals per histogram bin are Poisson distributed with mean Phi_sig or Phi_bkg.
    Imaging model in Sec. III-B relies on the standard Poisson SPAD model from [46], [47], including dead-time/pile-up effects.
  • domain assumption Laser photons arrive in a single bin, or in a known Gaussian pulse spanning few bins.
    Sec. III-B assumes no multipath or subsurface scattering in the forward model; non-ideal pulse shapes are acknowledged in hardware results.
  • domain assumption The depth prior locates the true histogram peak within the foveation window.
    The SBR gains in Eq. (6)-(7) and the memory savings in Alg. 1 assume the foveated M bins contain the signal peak; the paper calls this 'perfect foveation' in Sec. III-C2.
  • domain assumption Scene depths are coherent within quantized buckets or superpixels.
    Quantized sampling (Sec. V) and superpixel sampling (Sec. VII-B) assume all pixels in a bucket share the same depth, so sparse samples can represent each region.
  • standard math Brightness consistency holds for optical flow.
    Sec. VI warps previous-frame depths using optical flow vectors satisfying I(x+u dt, y+v dt, t+dt)=I(x,y,t).
  • domain assumption Future SPAD arrays will support per-pixel programmable gating and variable-resolution TDCs.
    Sec. III-A and VIII state the hardware does not yet exist; the paper proposes a speculative macropixel design in Fig. 10.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FoveaSPAD: Exploiting Depth Priors for Adaptive and Efficient Single-Photon 3D Imaging." pith.science (2026). https://pith.science/paper/F7JNSRZR

@misc{pith2026241202052,
  author       = {Pith},
  title        = {Pith review of: FoveaSPAD: Exploiting Depth Priors for Adaptive and Efficient Single-Photon 3D Imaging},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/F7JNSRZR}},
  note         = {Machine review of arXiv:2412.02052}
}
read the original abstract

Fast, efficient, and accurate depth-sensing is important for safety-critical applications such as autonomous vehicles. Direct time-of-flight LiDAR has the potential to fulfill these demands, thanks to its ability to provide high-precision depth measurements at long standoff distances. While conventional LiDAR relies on avalanche photodiodes (APDs), single-photon avalanche diodes (SPADs) are an emerging image-sensing technology that offer many advantages such as extreme sensitivity and time resolution. In this paper, we remove the key challenges to widespread adoption of SPAD-based LiDARs: their susceptibility to ambient light and the large amount of raw photon data that must be processed to obtain in-pixel depth estimates. We propose new algorithms and sensing policies that improve signal-to-noise ratio (SNR) and increase computing and memory efficiency for SPAD-based LiDARs. During capture, we use external signals to \emph{foveate}, i.e., guide how the SPAD system estimates scene depths. This foveated approach allows our method to ``zoom into'' the signal of interest, reducing the amount of raw photon data that needs to be stored and transferred from the SPAD sensor, while also improving resilience to ambient light. We show results both in simulation and also with real hardware emulation, with specific implementations achieving a 1548-fold reduction in memory usage, and our algorithms can be applied to newly available and future SPAD arrays.

Figures

Figures reproduced from arXiv: 2412.02052 by the authors.

Figure 1
Figure 1. Depth Prior Driven SPAD Depth Foveation: SPAD sensors suffer from a data bottleneck, since thousands of histogram bins are used to generate depth as shown in the top left. If fewer bins are used, this reduces depth resolution, as shown in the limited bins depth result. Our idea is to use additional information, such as a color image (Sec. IV, VII) or optical flow (Sec. VI), to foveate the SPAD bins. Therefore, for t… view at source ↗
Figure 2
Figure 2. Qualitative Comparison on NYUv2 Our memory and depth foveation techniques produce quality depth reconstructions with a fraction of the memory usage. Each row consists of the NYUv2 ground truth images, the monocular depth output from ZoeDepth, a simulated SPAD output with N′ bins, and our foveation techniques. The rows show different combinations of M and N′ , where M is the number of bins in the foveated histograms,… view at source ↗
Figure 3
Figure 3. Spatio-temporal foveation The first two columns display the scene’s color and ground truth depth. Using the quantized monocular depth in the third column, we select certain pixels in the fourth column. Processing only histograms at these locations with foveated windows generates results in the last column, indicating a 1548-fold reduction in memory usage. This is calculated by measuring memory allocation for full-re… view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Optical Flow Driven Foveation Here we see our optical flow driven SPAD foveation using the Carla simulator whose color and ground-truth depth are shown in the first two columns. Directly using optical flow, as shown in the third column, creates errors that propagate ov…
Figure 5
Figure 5. Figure 5: Hardware emulation results for scenes from Lindell et al. [4]. (Column 1) The Lindell dataset consists of monochrome images captured by a camera co-aligned with the SPAD sensor that captures photon data cubes. (Column 2) We obtain monocular depth maps from these monoch…
Figure 6
Figure 6. Figure 6: Hardware emulation results for scenes without co-aligned monochrome camera [5]. (Column 1) RGB images of the “face-vase” and “reindeer” scenes shown for visualization. (Column 2) A pseudo-intensity image is estimated by accumulating photon counts for each pixel. (Colum…
Figure 7
Figure 7. Figure 7: Additional Results: Depth Fovea. This figure demonstrates the application of the depth foveation technique described in Sec. IV to the Lindel dataset, along with the error correction technique presented in the supplementary material. A window size of M = 1/8 and a bin …
Figure 9
Figure 9. Figure 9: Effect of increasing background illumination. The conventional (non-foveated) depth map quality degrades more rapidly as background illumination increases. Using memory foveation allows reliable depth map recovery for the “deer” scene for a wider range of SBR levels. I…
Figure 8
Figure 8. Figure 8: Additional Results: Optical Flow and Quantization Spatio￾Temporal. This figure illustrates the application of the techniques described in Sec. VI and Sec. V to the Lindel dataset. The top portion showcases our optical flow algorithm on the ”roll” scene. The first colum…
Figure 10
Figure 10. Figure 10: Future pixel and array designs for foveated single-photon 3D imaging. (a) A speculative pixel design where individual SPADs are gated on or off based on thresholds set with respect to a linear ramp signal. Pixels only need to store the thresholds; the ramp signal is g…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

59 extracted references · 52 canonical work pages

  1. [1]

    Attention is all you need,

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” Advances in neural information processing systems , vol. 30, 2017

  2. [2]

    Grad-cam: Visual explanations from deep networks via gradient-based localization,

    R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad-cam: Visual explanations from deep networks via gradient-based localization,” in Proceedings of the IEEE international conference on computer vision , 2017, pp. 618–626

  3. [3]

    Depth from focusing and defocusing,

    Y . Xiong and S. Shafer, “Depth from focusing and defocusing,” in Proceedings of (CVPR) Computer Vision and Pattern Recognition , June 1993, pp. 68 – 73

  4. [4]

    Single-photon 3d imaging with deep sensor fusion,

    D. B. Lindell, M. O’Toole, and G. Wetzstein, “Single-photon 3d imaging with deep sensor fusion,” ACM Transactions on Graphics (ToG), vol. 37, no. 4, pp. 1–12, 2018

  5. [5]

    Compressive single-photon 3d cameras,

    F. Gutierrez-Barragan, A. Ingle, T. Seets, M. Gupta, and A. Vel- ten, “Compressive single-photon 3d cameras,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2022, pp. 17 854–17 864

  6. [6]

    A reconfigurable 3-d-stacked spad imager with in-pixel histogramming for flash lidar or high-speed time-of-flight imaging,

    S. W. Hutchings, N. Johnston, I. Gyongy, T. Al Abbas, N. A. Dutton, M. Tyler, S. Chan, J. Leach, and R. K. Henderson, “A reconfigurable 3-d-stacked spad imager with in-pixel histogramming for flash lidar or high-speed time-of-flight imaging,” IEEE Journal of Solid-State Circuits, vol. 54, no. 11, pp. 2947–2956, 2019

  7. [7]

    A 30-frames/s, 252 × 144 SPAD Flash LiDAR With 1728 Dual-Clock 48.8-ps TDCs, and Pixel-Wise Integrated Histogramming,

    C. Zhang, S. Lindner, I. M. Antolovi ´c, J. M. Pavia, M. Wolf, and E. Charbon, “A 30-frames/s, 252 × 144 SPAD Flash LiDAR With 1728 Dual-Clock 48.8-ps TDCs, and Pixel-Wise Integrated Histogramming,” IEEE Journal of Solid-State Circuits , vol. 54, no. 4, pp. 1137–1151, 2018

  8. [8]

    Sub-picosecond photon-efficient 3d imaging using single-photon sensors,

    F. Heide, S. Diamond, D. B. Lindell, and G. Wetzstein, “Sub-picosecond photon-efficient 3d imaging using single-photon sensors,” Scientific reports, vol. 8, no. 1, p. 17726, 2018

Show all 59 references
  1. [9]

    High-speed 3d sensing via hybrid-mode imaging and guided upsampling,

    I. Gyongy, S. W. Hutchings, A. Halimi, M. Tyler, S. Chan, F. Zhu, S. McLaughlin, R. K. Henderson, and J. Leach, “High-speed 3d sensing via hybrid-mode imaging and guided upsampling,”Optica, vol. 7, no. 10, pp. 1253–1260, 2020

  2. [10]

    Picosecond time-resolved imaging using spad cameras,

    G. Gariepy, J. Leach, R. Warburton, S. Chan, R. Henderson, and D. Faccio, “Picosecond time-resolved imaging using spad cameras,” in Emerging Imaging and Sensing Technologies , vol. 9992. SPIE, 2016, pp. 130–137

  3. [11]

    Learned compressive representations for single-photon 3d imaging,

    F. Gutierrez-Barragan, F. Mu, A. Ardelean, A. Ingle, C. Bruschini, E. Charbon, Y . Li, M. Gupta, and A. Velten, “Learned compressive representations for single-photon 3d imaging,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023, pp. 10 756–10 766

  4. [12]

    High-resolution depth profiling using a range-gated cmos spad quanta image sensor,

    X. Ren, P. W. Connolly, A. Halimi, Y . Altmann, S. McLaughlin, I. Gyongy, R. K. Henderson, and G. S. Buller, “High-resolution depth profiling using a range-gated cmos spad quanta image sensor,” Optics express, vol. 26, no. 5, pp. 5541–5557, 2018

  5. [13]

    A spad-based qvga image sensor for single-photon counting and quanta imaging,

    N. A. Dutton, I. Gyongy, L. Parmesan, S. Gnecchi, N. Calder, B. R. Rae, S. Pellegrini, L. A. Grant, and R. K. Henderson, “A spad-based qvga image sensor for single-photon counting and quanta imaging,” IEEE Transactions on Electron Devices , vol. 63, no. 1, pp. 189–196, 2015

  6. [14]

    A cmos spad line sensor with per- pixel histogramming tdc for time-resolved multispectral imaging,

    A. T. Erdogan, R. Walker, N. Finlayson, N. Krstaji ´c, G. Williams, J. Girkin, and R. Henderson, “A cmos spad line sensor with per- pixel histogramming tdc for time-resolved multispectral imaging,” IEEE Journal of Solid-State Circuits , vol. 54, no. 6, pp. 1705–1719, 2019

  7. [15]

    Single photon counting performance and noise analysis of cmos spad-based image sensors,

    N. A. Dutton, I. Gyongy, L. Parmesan, and R. K. Henderson, “Single photon counting performance and noise analysis of cmos spad-based image sensors,” Sensors, vol. 16, no. 7, p. 1122, 2016

  8. [16]

    A 240 × 160 3d-stacked spad dtof image sensor with rolling shutter and in-pixel histogram for mobile devices,

    C. Zhang, N. Zhang, Z. Ma, L. Wang, Y . Qin, J. Jia, and K. Zang, “A 240 × 160 3d-stacked spad dtof image sensor with rolling shutter and in-pixel histogram for mobile devices,” IEEE Open Journal of the Solid-State Circuits Society , vol. 2, pp. 3–11, 2021

  9. [17]

    Guided direct time-of-flight lidar using stereo cameras for enhanced laser power efficiency,

    F. Taneski, I. Gyongy, T. Al Abbas, and R. K. Henderson, “Guided direct time-of-flight lidar using stereo cameras for enhanced laser power efficiency,” Sensors, vol. 23, no. 21, p. 8943, 2023

  10. [18]

    Count-free single-photon 3d imaging with race logic,

    A. Ingle and D. Maier, “Count-free single-photon 3d imaging with race logic,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023

  11. [19]

    Adaptive gating for single- photon 3d imaging,

    R. Po, A. Pediredla, and I. Gkioulekas, “Adaptive gating for single- photon 3d imaging,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2022, pp. 16 354–16 363. 14

  12. [20]

    Lidar-flow: Dense scene flow estimation from sparse lidar and stereo images,

    R. Battrawy, R. Schuster, O. Wasenm ¨uller, Q. Rao, and D. Stricker, “Lidar-flow: Dense scene flow estimation from sparse lidar and stereo images,” in 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE, 2019, pp. 7762–7769

  13. [21]

    Estimat- ing depth from rgb and sparse sensing,

    Z. Chen, V . Badrinarayanan, G. Drozdov, and A. Rabinovich, “Estimat- ing depth from rgb and sparse sensing,” in Proceedings of the European Conference on Computer Vision (ECCV) , 2018, pp. 167–182

  14. [22]

    Sparse-to-dense: Depth prediction from sparse depth samples and a single image,

    F. Mal and S. Karaman, “Sparse-to-dense: Depth prediction from sparse depth samples and a single image,” in 2018 IEEE International Confer- ence on Robotics and Automation (ICRA) . IEEE, 2018, pp. 1–8

  15. [23]

    Sparse depth super resolution,

    J. Lu and D. Forsyth, “Sparse depth super resolution,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2015, pp. 2245–2253

  16. [24]

    Sparsity invariant cnns,

    J. Uhrig, N. Schneider, L. Schneider, U. Franke, T. Brox, and A. Geiger, “Sparsity invariant cnns,” in2017 International Conference on 3D Vision (3DV). IEEE, 2017, pp. 11–20

  17. [25]

    Atgv-net: Accurate depth super- resolution,

    G. Riegler, M. R ¨uther, and H. Bischof, “Atgv-net: Accurate depth super- resolution,” in European Conference on Computer Vision . Springer, 2016, pp. 268–284

  18. [26]

    Depth map super-resolution by deep multi-scale guidance,

    T.-W. Hui, C. C. Loy, and X. Tang, “Depth map super-resolution by deep multi-scale guidance,” in Proceedings of European Conference on Computer Vision (ECCV) , 2016

  19. [27]

    Sparse and noisy lidar completion with rgb guidance and uncertainty,

    W. Van Gansbeke, D. Neven, B. De Brabandere, and L. Van Gool, “Sparse and noisy lidar completion with rgb guidance and uncertainty,” arXiv preprint arXiv:1902.05356 , 2019

  20. [28]

    Gated2depth: Real-time dense lidar from gated images,

    T. Gruber, F. Julca-Aguilar, M. Bijelic, W. Ritter, K. Dietmayer, and F. Heide, “Gated2depth: Real-time dense lidar from gated images,”arXiv preprint arXiv:1902.04997, 2019

  21. [29]

    Saccadecam: Adaptive visual attention for monocular depth sensing,

    B. Tilmon and S. J. Koppal, “Saccadecam: Adaptive visual attention for monocular depth sensing,” in Proceedings of the IEEE/CVF Inter- national Conference on Computer Vision , 2021, pp. 6009–6018

  22. [30]

    Efficient pedestrian scanning by active scan lidar,

    T. Yamamoto, Y . Kawanishi, I. Ide, H. Murase, F. Shinmura, and D. Deguchi, “Efficient pedestrian scanning by active scan lidar,” in Advanced Image Technology (IWAIT), 2018 International Workshop on . IEEE, 2018, pp. 1–4

  23. [31]

    Directionally con- trolled time-of-flight ranging for mobile sensing platforms,

    Z. Tasneem, D. Wang, H. Xie, and S. J. Koppal, “Directionally con- trolled time-of-flight ranging for mobile sensing platforms,” in Robotics: Science and Systems , 2018

  24. [32]

    Deep adaptive lidar: End- to-end optimization of sampling and depth completion at low sampling rates,

    A. Bergman, D. Lindell, and G. Wetzstein, “Deep adaptive lidar: End- to-end optimization of sampling and depth completion at low sampling rates,” ICCP, 2020

  25. [33]

    Towards a mems-based adaptive lidar,

    F. Pittaluga, Z. Tasneem, J. Folden, B. Tilmon, A. Chakrabarti, and S. J. Koppal, “Towards a mems-based adaptive lidar,” in 2020 International Conference on 3D Vision (3DV) . IEEE, 2020, pp. 1216–1226

  26. [34]

    Foveated 3d graphics,

    B. Guenter, M. Finch, S. Drucker, D. Tan, and J. Snyder, “Foveated 3d graphics,” ACM transactions on Graphics (tOG) , vol. 31, no. 6, pp. 1–10, 2012

  27. [35]

    Latency requirements for foveated rendering in virtual reality,

    R. Albert, A. Patney, D. Luebke, and J. Kim, “Latency requirements for foveated rendering in virtual reality,” ACM Transactions on Applied Perception (TAP), vol. 14, no. 4, pp. 1–13, 2017

  28. [36]

    Luminance-contrast-aware foveated rendering,

    O. T. Tursun, E. Arabadzhiyska-Koleva, M. Wernikowski, R. Mantiuk, H.-P. Seidel, K. Myszkowski, and P. Didyk, “Luminance-contrast-aware foveated rendering,” ACM Transactions on Graphics (TOG) , vol. 38, no. 4, pp. 1–14, 2019

  29. [37]

    The light field stereo- scope

    F.-C. Huang, D. P. Luebke, and G. Wetzstein, “The light field stereo- scope.” in SIGGRAPH emerging technologies , 2015, pp. 24–1

  30. [38]

    Perceptually-guided foveation for light field displays,

    Q. Sun, F.-C. Huang, J. Kim, L.-Y . Wei, D. Luebke, and A. Kaufman, “Perceptually-guided foveation for light field displays,” ACM Transac- tions on Graphics (TOG) , vol. 36, no. 6, pp. 1–13, 2017

  31. [39]

    Perceptually-based foveated virtual reality,

    A. Patney, J. Kim, M. Salvi, A. Kaplanyan, C. Wyman, N. Benty, A. Lefohn, and D. Luebke, “Perceptually-based foveated virtual reality,” in ACM SIGGRAPH 2016 emerging technologies , 2016, pp. 1–2

  32. [40]

    3d-kernel foveated rendering for light fields,

    X. Meng, R. Du, J. F. JaJa, and A. Varshney, “3d-kernel foveated rendering for light fields,” IEEE transactions on visualization and computer graphics, vol. 27, no. 8, pp. 3350–3360, 2020

  33. [41]

    A sketching framework for reduced data transfer in photon counting lidar,

    M. Sheehan, J. Tachella, and M. Davies, “A sketching framework for reduced data transfer in photon counting lidar,” IEEE Transactions on Computational Imaging, vol. 7, p. 989–1004, 2021. [Online]. Available: http://dx.doi.org/10.1109/TCI.2021.3113495

  34. [42]

    First arrival differential lidar,

    T. Zhang, M. J. White, A. Dave, S. Ghajari, A. Raghuram, A. C. Molnar, and A. Veeraraghavan, “First arrival differential lidar,” in 2022 IEEE International Conference on Computational Photography (ICCP), 2022, pp. 1–12

  35. [43]

    A differential spad array architecture in 0.18 um cmos for hdr imaging,

    M. White, S. Ghajari, T. Zhang, A. Dave, A. Veeraraghavan, and A. Molnar, “A differential spad array architecture in 0.18 um cmos for hdr imaging,” in 2022 IEEE International Symposium on Circuits and Systems (ISCAS), 2022, pp. 292–296

  36. [44]

    Histogram-less lidar through spad response linearization,

    A. Tontini, S. Mazzucchi, R. Passerone, N. Broseghini, and L. Gas- parini, “Histogram-less lidar through spad response linearization,” IEEE Sensors Journal, vol. PP, pp. 1–1, 01 2023

  37. [45]

    End-to-end learned, optically coded super-resolution spad camera,

    Q. Sun, J. Zhang, X. Dun, B. Ghanem, Y . Peng, and W. Heidrich, “End-to-end learned, optically coded super-resolution spad camera,” ACM Trans. Graph. , vol. 39, no. 2, mar 2020. [Online]. Available: https://doi.org/10.1145/3372261

  38. [46]

    Asynchronous single-photon 3d imaging,

    A. Gupta, A. Ingle, and M. Gupta, “Asynchronous single-photon 3d imaging,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 7909–7918

  39. [47]

    Photon-flooded single- photon 3d cameras,

    A. Gupta, A. Ingle, A. Velten, and M. Gupta, “Photon-flooded single- photon 3d cameras,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2019, pp. 6770–6779

  40. [48]

    itof2dtof: A robust and flexible representation for data-driven time-of- flight imaging,

    F. Gutierrez-Barragan, H. Chen, M. Gupta, A. Velten, and J. Gu, “itof2dtof: A robust and flexible representation for data-driven time-of- flight imaging,” IEEE Transactions on Computational Imaging , vol. 7, pp. 1205–1214, 2021

  41. [49]

    Indoor segmen- tation and support inference from rgbd images,

    P. K. Nathan Silberman, Derek Hoiem and R. Fergus, “Indoor segmen- tation and support inference from rgbd images,” in ECCV, 2012

  42. [50]

    Are we ready for autonomous driving? the kitti vision benchmark suite,

    A. Geiger, P. Lenz, and R. Urtasun, “Are we ready for autonomous driving? the kitti vision benchmark suite,” in Conference on Computer Vision and Pattern Recognition (CVPR) , 2012

  43. [51]

    Zoedepth: Zero-shot transfer by combining relative and metric depth,

    S. F. Bhat, R. Birkl, D. Wofk, P. Wonka, and M. M ¨uller, “Zoedepth: Zero-shot transfer by combining relative and metric depth,” 2023. [Online]. Available: https://arxiv.org/abs/2302.12288

  44. [52]

    Caspi: col- laborative photon processing for active single-photon imaging,

    J. Lee, A. Ingle, J. V . Chacko, K. W. Eliceiri, and M. Gupta, “Caspi: col- laborative photon processing for active single-photon imaging,” Nature Communications, vol. 14, no. 1, p. 3158, 2023

  45. [53]

    Spad-based flash lidar sensor with high ambient light rejection for automotive applications,

    M. Beer, O. M. Schrey, J. F. Haase, J. Ruskowski, W. Brockherde, B. J. Hosticka, and R. Kokozinski, “Spad-based flash lidar sensor with high ambient light rejection for automotive applications,” inQuantum Sensing and Nano Electronics and Photonics XV , vol. 10540. SPIE, 2018, ...

  46. [54]

    Carla: An open urban driving simulator,

    A. Dosovitskiy, G. Ros, F. Codevilla, A. Lopez, and V . Koltun, “Carla: An open urban driving simulator,” in Conference on robot learning . PMLR, 2017, pp. 1–16

  47. [55]

    Linospad: a compact linear spad camera system with 64 fpga-based tdc modules for versatile 50 ps resolution time-resolved imaging,

    S. Burri, C. Bruschini, and E. Charbon, “Linospad: a compact linear spad camera system with 64 fpga-based tdc modules for versatile 50 ps resolution time-resolved imaging,” Instruments, vol. 1, no. 1, p. 6, 2017

  48. [56]

    Superpixels and polygons using simple non-iterative clustering,

    R. Achanta and S. Susstrunk, “Superpixels and polygons using simple non-iterative clustering,” in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017

  49. [57]

    Computational imaging spad cameras,

    A. Ardelean, “Computational imaging spad cameras,” Ph.D. dissertation, EPFL, 2023

  50. [58]

    Sodacam: Software-defined cameras via single-photon imag- ing,

    V . Sundar, A. Ardelean, T. Swedish, C. Bruschini, E. Charbon, and M. Gupta, “Sodacam: Software-defined cameras via single-photon imag- ing,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 8165–8176

  51. [59]

    Low-cost spad sensing for non-line-of-sight tracking, material classification and depth imaging,

    C. Callenberg, Z. Shi, F. Heide, and M. B. Hullin, “Low-cost spad sensing for non-line-of-sight tracking, material classification and depth imaging,” ACM Transactions on Graphics (TOG) , vol. 40, no. 4, pp. 1–12, 2021

Pith tools

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