Paper Info Reviews Author Feedback Meta-Review Back to top
List of Papers Browse by Subject Areas Author List

Abstract

A single-pixel camera is a spatial-multiplexing device that reconstructs an image from a sequence of projections of the scene onto some patterns. This architecture is used, for example, to assist neurosurgery with hyperspectral imaging. However, capturing dynamic scenes is very challenging: as the different projections measure different frames of the scene, standard reconstruction approaches suffer from strong motion artifacts. This paper presents a general framework to reconstruct a moving scene with two main contributions. First, we extend the field of view of the camera beyond that defined by the spatial light modulator, which dramatically reduces the model mismatch. Second, we propose to build the dynamic system matrix without warping the patterns, effectively dismissing discretization errors. Numerical experiments show that both our contributions are necessary for an artifact-free reconstruction. The influence of a reduced measured set, robustness to noise and to motion errors were also evaluated.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2024/paper/1903_paper.pdf

SharedIt Link: https://rdcu.be/dV5C7

SpringerLink (DOI): https://doi.org/10.1007/978-3-031-72104-5_27

Supplementary Material: https://papers.miccai.org/miccai-2024/supp/1903_supp.pdf

Link to the Code Repository

https://github.com/openspyrit/spyrit

Link to the Dataset(s)

N/A

BibTex

@InProceedings{Mai_Dynamic_MICCAI2024,
        author = { Maitre, Thomas and Bretin, Elie and Phan, Romain and Ducros, Nicolas and Sdika, Michaël},
        title = { { Dynamic Single-Pixel Imaging on an Extended Field of View without Warping the Patterns } },
        booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2024},
        year = {2024},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 15007},
        month = {October},
        page = {275 -- 284}
}

Reviews

Review #1

  • Please describe the contribution of the paper

    This paper presents a general framework to reconstruct a moving scene. The contributions include: a new method for dynamic single pixel imaging on an extended FOV and a new discretization for the dynamic forward model that removes artifacts.

  • Please list the main strengths of the paper; you should write about a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
    1. Solve the challenge of capturing dynamic scenes by single-pixel camera
    2. writing is ok, paper organization is clear.
  • Please list the main weaknesses of the paper. Please provide details, for instance, if you think a method is not novel, explain why and provide a reference to prior work.
    1. it lacks of an overview Figure to clearly show the technique of the whole solution.
    2. Lack of descriptions on state of the art comparison method.
    3. datasets/test sequences are not clearly given.
    4. Metric is not enough, only PSNR and SSIM are shown, subjective quality metric, e.g., LPIPS, VFID might be also useful.
  • Please rate the clarity and organization of this paper

    Satisfactory

  • Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.

    The submission does not provide sufficient information for reproducibility.

  • Do you have any additional comments regarding the paper’s reproducibility?

    N/A

  • Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review. Pay specific attention to the different assessment criteria for the different paper categories (MIC, CAI, Clinical Translation of Methodology, Health Equity): https://conferences.miccai.org/2024/en/REVIEWER-GUIDELINES.html

    This paper solve the dynamic problem on an extended FOV and build the dynamic forward model without warping the light patterns, allowing for an artifact-free reconstruction of the scene.

  • Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making

    Weak Reject — could be rejected, dependent on rebuttal (3)

  • Please justify your recommendation. What were the major factors that led you to your overall score for this paper?

    performance, writting, real world applications

  • Reviewer confidence

    Somewhat confident (2)

  • [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed

    N/A

  • [Post rebuttal] Please justify your decision

    N/A



Review #2

  • Please describe the contribution of the paper

    The paper presents a novel methodology for dynamic scene reconstruction using a single-pixel camera, highlighting two key improvements: extending the field of view and constructing a dynamic system matrix without warping the patterns. Demonstrated with simulated data, this innovation addresses the challenges in imaging for moving conditions, such as in neurosurgery, reducing artifacts caused by model mismatches and discretization errors.

  • Please list the main strengths of the paper; you should write about a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
    1. The development of a novel dynamic forward model that avoids the traditional warping of patterns is significant. The extended reconstruction FOV is also a smart way to tackle the problem of boundary occlusion.
    2. Rigorous numerical experiments compare the new method with multiple regularization techniques, demonstrating its robustness against noise and motion estimation errors.
  • Please list the main weaknesses of the paper. Please provide details, for instance, if you think a method is not novel, explain why and provide a reference to prior work.
    • Although the paper compares various regularization methods, all experiments are conducted on simulated data with Sine motion and Poisson noise. It remains uncertain how effectively the proposed method will perform on real-world data.
  • Please rate the clarity and organization of this paper

    Excellent

  • Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.

    The authors claimed to release the source code and/or dataset upon acceptance of the submission.

  • Do you have any additional comments regarding the paper’s reproducibility?

    N/A

  • Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review. Pay specific attention to the different assessment criteria for the different paper categories (MIC, CAI, Clinical Translation of Methodology, Health Equity): https://conferences.miccai.org/2024/en/REVIEWER-GUIDELINES.html

    Generally speaking, this paper is already very well-established. Including a discussion on potential limitations, challenges, and avenues for future research could provide a more well-rounded view of this study. Demonstrating the effectiveness of this method with real-world data instead of simulated data would be great. However, it’s worth noting that including more experiments in the rebuttal may not be encouraged under the MICCAI guideline.

  • Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making

    Accept — should be accepted, independent of rebuttal (5)

  • Please justify your recommendation. What were the major factors that led you to your overall score for this paper?

    The paper is recommended for acceptance due to its innovative approach to dynamic scene reconstruction using a single-pixel camera, particularly in extending the field of view and eliminating the need for pattern warping. These contributions meaningfully address key challenges in neurosurgical single-pixel imaging. While experiments use simulated data, the robust methodology and clear potential for clinical applications provide a compelling case for the paper’s value in the field.

  • Reviewer confidence

    Confident but not absolutely certain (3)

  • [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed

    N/A

  • [Post rebuttal] Please justify your decision

    N/A



Review #3

  • Please describe the contribution of the paper

    The manuscript offers means to add motion compensation to the measurement process in a single-pixel imaging system to enable dynamic imaging. Furthermore, the work demonstrates that the field of view in the reconstruction domain must be extended to avoid artifacts. The results show clear and significant computational and reconstructed image quality improvements when compared to the state of the art benchmark.

  • Please list the main strengths of the paper; you should write about a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
    • The mathematical formulation is concise and clear.
    • The results are fairly comprehensive given the limited space.
    • The results show significant improvement over the state of the art.
  • Please list the main weaknesses of the paper. Please provide details, for instance, if you think a method is not novel, explain why and provide a reference to prior work.
    • The results of the “Robustness to motion errors” experiment are extremely important to validate the approach’s robustness. However, the results are somewhat difficult to interpret and too simple. By difficult to interpret, I mean that (if I am not mistaken) the parameters a and T of the synthetic motion are not documented and it is not easy assess the significance of the motion. By too simple, the comparison is made to an assumption of a static scene and I believe a more compelling and realistic result would be use a motion estimation approach and investigate the error.
    • There are some typos in the references, e.g., pet -> PET, ct -> CT, mri -> MRI.
  • Please rate the clarity and organization of this paper

    Excellent

  • Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.

    The authors claimed to release the source code and/or dataset upon acceptance of the submission.

  • Do you have any additional comments regarding the paper’s reproducibility?

    Nothing more than the comments in the “main weaknesses” above.

  • Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review. Pay specific attention to the different assessment criteria for the different paper categories (MIC, CAI, Clinical Translation of Methodology, Health Equity): https://conferences.miccai.org/2024/en/REVIEWER-GUIDELINES.html
    • Could a basis tailored to the expected motion in (6) be developed to refine the approach?
    • Is there any significance to the dips in the blue curved in Fig. 3 (a) and (b)?
  • Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making

    Accept — should be accepted, independent of rebuttal (5)

  • Please justify your recommendation. What were the major factors that led you to your overall score for this paper?

    Overall, the method is well justified and mathematically developed. The results are fairly comprehensive and provide evidence of a significant improvement over the state of the art. For these reasons, I think the paper should be considered for acceptance.

  • Reviewer confidence

    Confident but not absolutely certain (3)

  • [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed

    N/A

  • [Post rebuttal] Please justify your decision

    N/A



Author Feedback

Reviewer 1:

  • Metrics are not enough: Thank you for the advice, we will consider showing more metrics in an extended journal version.

Reviewer 3 and 4:

  • Motion is too simple / method only tested on synthetic data: Reconstruction of a simple object under motion on real phantom data has been reported in [11] for an object constantly within the FOV. We are in the process of acquiring real, experimental data with an object entering/leaving the FOV and hopefully will be able to present the results soon to validate the new contributions in real condition.

Reviewer 4:

  • parameters a and T of the synthetic motion are not documented and it is not easy assess the significance of the motion : Thank you. For these experiments, they were fixed to the values of a = 0.2 and T = 1000 ms and we considered an acquisition of 2000 ms to match the conditions of [11]. The missing values will be added to the camera ready paper and influence of these parameters will be studied in an extended version of the paper.

  • robustness to motion errors makes “an assumption of a static scene and I believe a more compelling and realistic result would be use a motion estimation approach and investigate the error”: The motion field used in the reconstruction method has been already estimated with the motion estimation approach proposed in [18] in this work (last paragraph of the Results section). There is no assumption of a static scene.

  • There are some typos in the references, e.g., pet -> PET, ct -> CT, mri -> MRI. Thank you, this will be corrected.



Meta-Review

Meta-review not available, early accepted paper.



back to top