- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
- This paper try to solve the image deblurring task using the proposed method called UniDeblur.
- The main idea of the paper is replacing the residual connection in the Unet with the Fourier Convolution Learning (FCL).
- Multiple datasets are used to train the network to avoid reduce the domain gap between datasets, the OOD testing proved the effectiveness of this training. Thus, authors also claim that the proposed method is a "unified motion deblurring framework"
Critical strengths
- The kernel shown in the Figure 6 correlated with what adaptive filter learned is convincing.
- simple and easy-to-understand ideas and methods. Fourier transform convert the convolution to element-wise multiplication, which avoid the OOM when the convolution kernel size is too large.
- The adaptive filter mimic the dynamic convolution that adapt the generated kernels for different images, which is well-motivated.
- The experiments are broad. They cover a lot of datasets with OOD testing.
- All compared methods are retrained on new datasets instead of naively copying values from the original papers.
- FCL is general, it can be plug-and-play into different architectures.
Critical weaknesses
- One of my biggest concern is that only the convolution (spatial-invariant blur) can be represent with the multiplication in the Fourier domain, a lot of the classic image deblurring datasets are handling spatial-variant motion blur like GoPro, HIDE, REDS, etc. I think the key assumption is not correct in my opinion. Spatial-variant blur can only be represented as matrix multiplication instead of the convolution.
- Though Figure 6 shows the learned kernels, but a more convincing experiments could be: with known blur kernels, what will the learned filters look like. Will they look similar?
- I know it makes sense to use no-reference image quality metrics to evaluate the OOD because the gt is imperfect, a user study will be preferred since these no-reference image evaluators are far from perfect, sometimes they prefer noisy than clean images.
- The dataset pruning is a bit unconvincing. I know it makes sense somehow, but pruning could also removed some hard cases that cannot be resolved using the proposed methods, for example spatial-variant blur. If so, we cannot say the proposed method is "unified".
- Contribution 2: reformulates complex spatial kernel prediction into efficient frequency-domain multiplication. I feel this has been done in tons of previous works, for example: @inproceedings{xintm2025DeepRFTv2, title = {DeepRFTv2: Fourier Kernel Estimation for Image Deblurring}, author = {Xintian Mao, JIansheng Wang, Xingran Xie, Qingli Li and Yan Wang}, booktitle = {arXiv}, year = {2025} }
Questions for authors
- How did authors obtain and visualize the kernels in Figure 6 and 7.
- How does FCL handle spatial-variant blur? This is critical!!!
- Any details about the data prunning?
- How sensitive are the retrained models on the pruned data.
- Have you done any user study?
2 comments
- My spatial-variant kernel concern is not resolved. I know "neural networks inherently possess the ability to handle spatially varying motion blur", but multiplying by a fixed Fourier filter still corresponds to spatially invariant convolution within the region represented by that filter. How do different patches learn different motion kernels? It would make more sense if the patch-based FCL is used instead of image-based. And kernels shown in Figure 6 and 7 are still global kernels instead of local kernels.
- I do not think user study is useless. It can complement the quantitative metrics instead of replace them.
- How sensitive the results are to pruning is not explained. "Nearly all models can effectively handle motion blur, but fail to process defocus blur." does not have experimental supports.
- I am not saying the authors plagiarized DeepRFTv2, I just meant the Fourier kernels idea is widely used in this topic. So the distinguish of this work and other works is still unclear to me.
Response to Reviewer jC4n
- As explained in our previous response, at test time, we apply FCL to patchified features rather than to the full image.
- Thanks for this suggestion, and we will consider it.
- Because we have not yet devised a suitable experiment to demonstrate how sensitive the results are to pruning.
- We accept the your judgment and plan to withdraw this submission. We realize that our writing led to the reviewers' confusion, and we will address these issues in an updated version.