- Rating
- 1
- Confidence
- 4
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
This paper proposes a unified view of self-supervised methods for image denoising, highlighting the role played by the so-called trace term that needs to be close to zero. Based on this trace-decomposition, the authors construct a two-stage single-image denoiser named LoTA-N2N that penalizes the trace term. They show results on synthetic Gaussian, Poisson and mixed Poisson-Gaussian noise.
Critical strengths
This paper tries to give a unified framework of self-supervised methods for image denoising, advocating all methods are linked by the fact that the so-called trace term is zero (or close to zero) whatever the parameters are (section 3).
Critical weaknesses
Major:
-
I don't believe in the numerical values in Table 1. For example, the PSNR values of BM3D and DnCNN are under-estimated when compared to the ones given in [a]. The gap between the supervised approach (DnCNN here) and N2N should be close to zero, see for example [b]. Why does the performance of DnCNN collapse on Poisson noise when evaluating on McMaster18 instead on Kodak24? All of this is aberrant. Please clarify. Moreover, it is well-established that BM3D outperforms ZS-N2N on both Gaussian noise and Poisson noise with appropriate variance-stabilizing transformation, see for example [c]. Given the close scores between ZS-N2N and your method, I cannot believe that LoTA-N2N beats BM3D, contrary to what Table 1 suggests. Same holds for Table 8 where the performance of Self2Self is under-estimated (see [c, d]). Indeed, the performance of Self2Self should be way higher than ZSN2N on Gaussian noise, not 2dB lower... In summary, the clear aberrant PSNR values in Table 1 and 8 are very suspicious to me and the authors have lost my trust. This is a sufficient reason for me to reject this paper.
-
The method is very close to ZSN2N both in terms of construction and experimental results (see Table 1 and Figure 3 and 10 where the same denoising artifacts are produced). The problem is that ZSN2N is already far from being state-of-the-art, see [c] for example.
-
The paper is not really well written and hard to follow. For example, some notations are used without being defined, or sometimes are defined further, etc.
[a] Zhang, K., Zuo, W., & Zhang, L. FFDNet: Toward a fast and flexible solution for CNN-based image denoising. IEEE TIP, 2018.
[b] Wang, Z., Liu, J., Li, G., & Han, H. Blind2unblind: Self-supervised image denoising with visible blind spots. CVPR, 2022.
[c] Herbreteau, S., & Unser M. Self-Calibrated Variance-Stabilizing Transformations for Real-World Image Denoising. ICCV, 2025.
[d] Quan, Y., Chen, M., Pang, T., & Ji, H. Self2self with dropout: Learning self-supervised denoising from single image. CVPR, 2020.
Minor:
- Line 150, you define as a self-supervised input. Do you mean the noisy input image? If so, I recommend renaming it as which is the common notation in image denoising. But, from what I understand, you want to encompass recorrupted noisy images as in R2R with for example. The problem is that, in this case, the supervised loss is not =
but rather , so that the self-supervised and supervised minimization do not agree whenever , contrary to your claim line 158. Please clarify.
-
The trace has a precise meaning in linear algebra (sum of the diagonal elements of a square matrix). Since you use the same denomination to denote the quantity defined in (3), I recommend warning the reader about it to avoid potential confusion.
-
Are , , matrices? Generally, an image is interpreted as a vector. Therefore, you do not need the Frobenius norm and Frobenius inner product in equations (1)-(3).
-
Line 834, you say that " and average the anti-diagonal and diagonal pixels of each non-overlapping 2×2 block". However, this does not seem to be the case in Figure 2. Can you please clarify?
-
In equation (4), you do not define the quantities , , etc... As these notations are more or less introduced in the following paragraphs, I recommend to transform equation (4) into a Figure instead for improved clarity.
-
In Figure 2, MSE vs is written twice in Stage 1. I think this is a typo.
-
Appendix C should probably be moved to the main paper.
-
I don't understand equation (30). What is ? Is ?
-
Line 873, what is exactly ? Why does it depend on ? What is ?
-
Please use correcty \citet and \citep when citing in your Latex file.
Questions for authors
- Why do you need differentiation and expectation to be interchanged to write equation (8)? Isn't it a trivial consequence of equation (2)? If I understand correctly, when for all , then and there is no gradient-gap (of course in practice, it is not true since we use instead an empirical approximation of the expected value, see [e] for example).
[e] Klug, T., Atik, D., & Heckel, R. Analyzing the sample complexity of self-supervised image reconstruction methods. NeurIPS, 2023.
- As mentioned line 249, "because neighboring clean pixels may differ and their noise may be correlated, the resulting pair need not satisfy N2N assumptions". What solution do you propose for correlated noise?