Seongmin Hong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Seongmin Hong has a total of 113 co-authors. Here are the top co-authors by number of shared publications:
- Se Young Chun
Seoul National University
7 shared publications - Jaehyeok Bae
Dept. of Electrical and Computer Engineering
2 shared publications - Jongho Lee
Department of Electrical and Computer Engineering Seoul National University Seoul South Korea
2 shared publications - Abdel Younes
Synaptics Europe
1 shared publication Andrew Wei
TCL Corporate Research
1 shared publication- Andrey Ignatov
ETH Zurich
1 shared publication
Latest Publications
- 5 November 2025
- Published by Elsevier in Image and Vision Computing
- Vol. 165
Fidelity-preserving zero-shot diffusion models for highly ill-posed inverse problems in lensless imaging
- 5 November 2025
- Published by Elsevier in Image and Vision Computing
- Vol. 165
Fidelity-preserving zero-shot diffusion models for highly ill-posed inverse problems in lensless imaging
- 21 November 2024
- Published by Springer Nature
- p. 307-322
Adaptive Selection of Sampling-Reconstruction in Fourier Compressed Sensing
- 21 November 2024
- Published by Springer Nature
- p. 307-322
Adaptive Selection of Sampling-Reconstruction in Fourier Compressed Sensing
- Open Access
Adaptive Selection of Sampling-Reconstruction in Fourier Compressed Sensing
Compressed sensing (CS) has emerged to overcome the inefficiency of Nyquist sampling. However, traditional optimization-based reconstruction is slow and can not yield an exact image in practice. Deep learning-based reconstruction has been a promising alternative to optimization-based reconstruction, outperforming it in accuracy and computation speed. Finding an efficient sampling method with deep learning-based reconstruction, especially for Fourier CS remains a challenge. Existing joint optimization of sampling-reconstruction works ($\mathcal{H}_1$) optimize the sampling mask but have low potential as it is not adaptive to each data point. Adaptive sampling ($\mathcal{H}_2$) has also disadvantages of difficult optimization and Pareto sub-optimality. Here, we propose a novel adaptive selection of sampling-reconstruction ($\mathcal{H}_{1.5}$) framework that selects the best sampling mask and reconstruction network for each input data. We provide theorems that our method has a higher potential than $\mathcal{H}_1$ and effectively solves the Pareto sub-optimality problem in sampling-reconstruction by using separate reconstruction networks for different sampling masks. To select the best sampling mask, we propose to quantify the high-frequency Bayesian uncertainty of the input, using a super-resolution space generation model. Our method outperforms joint optimization of sampling-reconstruction ($\mathcal{H}_1$) and adaptive sampling ($\mathcal{H}_2$) by achieving significant improvements on several Fourier CS problems.Open AccessAdaptive Selection of Sampling-Reconstruction in Fourier Compressed Sensing
Compressed sensing (CS) has emerged to overcome the inefficiency of Nyquist sampling. However, traditional optimization-based reconstruction is slow and can not yield an exact image in practice. Deep learning-based reconstruction has been a promising alternative to optimization-based reconstruction, outperforming it in accuracy and computation speed. Finding an efficient sampling method with deep learning-based reconstruction, especially for Fourier CS remains a challenge. Existing joint optimization of sampling-reconstruction works ($\mathcal{H}_1$) optimize the sampling mask but have low potential as it is not adaptive to each data point. Adaptive sampling ($\mathcal{H}_2$) has also disadvantages of difficult optimization and Pareto sub-optimality. Here, we propose a novel adaptive selection of sampling-reconstruction ($\mathcal{H}_{1.5}$) framework that selects the best sampling mask and reconstruction network for each input data. We provide theorems that our method has a higher potential than $\mathcal{H}_1$ and effectively solves the Pareto sub-optimality problem in sampling-reconstruction by using separate reconstruction networks for different sampling masks. To select the best sampling mask, we propose to quantify the high-frequency Bayesian uncertainty of the input, using a super-resolution space generation model. Our method outperforms joint optimization of sampling-reconstruction ($\mathcal{H}_1$) and adaptive sampling ($\mathcal{H}_2$) by achieving significant improvements on several Fourier CS problems. - Open Access
On Exact Inversion of DPM-Solvers
Diffusion probabilistic models (DPMs) are a key component in modern generative models. DPM-solvers have achieved reduced latency and enhanced quality significantly, but have posed challenges to find the exact inverse (i.e., finding the initial noise from the given image). Here we investigate the exact inversions for DPM-solvers and propose algorithms to perform them when samples are generated by the first-order as well as higher-order DPM-solvers. For each explicit denoising step in DPM-solvers, we formulated the inversions using implicit methods such as gradient descent or forward step method to ensure the robustness to large classifier-free guidance unlike the prior approach using fixed-point iteration. Experimental results demonstrated that our proposed exact inversion methods significantly reduced the error of both image and noise reconstructions, greatly enhanced the ability to distinguish invisible watermarks and well prevented unintended background changes consistently during image editing. Project page: \url{https://smhongok.github.io/inv-dpm.html}.
Open AccessOn Exact Inversion of DPM-Solvers
Diffusion probabilistic models (DPMs) are a key component in modern generative models. DPM-solvers have achieved reduced latency and enhanced quality significantly, but have posed challenges to find the exact inverse (i.e., finding the initial noise from the given image). Here we investigate the exact inversions for DPM-solvers and propose algorithms to perform them when samples are generated by the first-order as well as higher-order DPM-solvers. For each explicit denoising step in DPM-solvers, we formulated the inversions using implicit methods such as gradient descent or forward step method to ensure the robustness to large classifier-free guidance unlike the prior approach using fixed-point iteration. Experimental results demonstrated that our proposed exact inversion methods significantly reduced the error of both image and noise reconstructions, greatly enhanced the ability to distinguish invisible watermarks and well prevented unintended background changes consistently during image editing. Project page: \url{https://smhongok.github.io/inv-dpm.html}.
- Open Access
- 26 June 2023
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 37 (10) , 12225-12233
Neural Diffeomorphic Non-uniform B-spline Flows
Normalizing flows have been successfully modeling a complex probability distribution as an invertible transformation of a simple base distribution. However, there are often applications that require more than invertibility. For instance, the computation of energies and forces in physics requires the second derivatives of the transformation to be well-defined and continuous. Smooth normalizing flows employ infinitely differentiable transformation, but with the price of slow non-analytic inverse transforms. In this work, we propose diffeomorphic non-uniform B-spline flows that are at least twice continuously differentiable while bi-Lipschitz continuous, enabling efficient parametrization while retaining analytic inverse transforms based on a sufficient condition for diffeomorphism. Firstly, we investigate the sufficient condition for C(k-2)-diffeomorphic non-uniform kth-order B-spline transformations. Then, we derive an analytic inverse transformation of the non-uniform cubic B-spline transformation for neural diffeomorphic non-uniform B-spline flows. Lastly, we performed experiments on solving the force matching problem in Boltzmann generators, demonstrating that our C2-diffeomorphic non-uniform B-spline flows yielded solutions better than previous spline flows and faster than smooth normalizing flows. Our source code is publicly available at https://github.com/smhongok/Non-uniform-B-spline-Flow.
- 26 June 2023
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 37 (10) , 12225-12233
Open AccessNeural Diffeomorphic Non-uniform B-spline Flows
Normalizing flows have been successfully modeling a complex probability distribution as an invertible transformation of a simple base distribution. However, there are often applications that require more than invertibility. For instance, the computation of energies and forces in physics requires the second derivatives of the transformation to be well-defined and continuous. Smooth normalizing flows employ infinitely differentiable transformation, but with the price of slow non-analytic inverse transforms. In this work, we propose diffeomorphic non-uniform B-spline flows that are at least twice continuously differentiable while bi-Lipschitz continuous, enabling efficient parametrization while retaining analytic inverse transforms based on a sufficient condition for diffeomorphism. Firstly, we investigate the sufficient condition for C(k-2)-diffeomorphic non-uniform kth-order B-spline transformations. Then, we derive an analytic inverse transformation of the non-uniform cubic B-spline transformation for neural diffeomorphic non-uniform B-spline flows. Lastly, we performed experiments on solving the force matching problem in Boltzmann generators, demonstrating that our C2-diffeomorphic non-uniform B-spline flows yielded solutions better than previous spline flows and faster than smooth normalizing flows. Our source code is publicly available at https://github.com/smhongok/Non-uniform-B-spline-Flow.
- Open Access
Neural Diffeomorphic Non-uniform B-spline Flows
Normalizing flows have been successfully modeling a complex probability distribution as an invertible transformation of a simple base distribution. However, there are often applications that require more than invertibility. For instance, the computation of energies and forces in physics requires the second derivatives of the transformation to be well-defined and continuous. Smooth normalizing flows employ infinitely differentiable transformation, but with the price of slow non-analytic inverse transforms. In this work, we propose diffeomorphic non-uniform B-spline flows that are at least twice continuously differentiable while bi-Lipschitz continuous, enabling efficient parametrization while retaining analytic inverse transforms based on a sufficient condition for diffeomorphism. Firstly, we investigate the sufficient condition for Ck-2-diffeomorphic non-uniform kth-order B-spline transformations. Then, we derive an analytic inverse transformation of the non-uniform cubic B-spline transformation for neural diffeomorphic non-uniform B-spline flows. Lastly, we performed experiments on solving the force matching problem in Boltzmann generators, demonstrating that our C2-diffeomorphic non-uniform B-spline flows yielded solutions better than previous spline flows and faster than smooth normalizing flows. Our source code is publicly available at https://github.com/smhongok/Non-uniform-B-spline-Flow.
Open AccessNeural Diffeomorphic Non-uniform B-spline Flows
Normalizing flows have been successfully modeling a complex probability distribution as an invertible transformation of a simple base distribution. However, there are often applications that require more than invertibility. For instance, the computation of energies and forces in physics requires the second derivatives of the transformation to be well-defined and continuous. Smooth normalizing flows employ infinitely differentiable transformation, but with the price of slow non-analytic inverse transforms. In this work, we propose diffeomorphic non-uniform B-spline flows that are at least twice continuously differentiable while bi-Lipschitz continuous, enabling efficient parametrization while retaining analytic inverse transforms based on a sufficient condition for diffeomorphism. Firstly, we investigate the sufficient condition for Ck-2-diffeomorphic non-uniform kth-order B-spline transformations. Then, we derive an analytic inverse transformation of the non-uniform cubic B-spline transformation for neural diffeomorphic non-uniform B-spline flows. Lastly, we performed experiments on solving the force matching problem in Boltzmann generators, demonstrating that our C2-diffeomorphic non-uniform B-spline flows yielded solutions better than previous spline flows and faster than smooth normalizing flows. Our source code is publicly available at https://github.com/smhongok/Non-uniform-B-spline-Flow.
- Open Access
On the Robustness of Normalizing Flows for Inverse Problems in Imaging
Conditional normalizing flows can generate diverse image samples for solving inverse problems. Most normalizing flows for inverse problems in imaging employ the conditional affine coupling layer that can generate diverse images quickly. However, unintended severe artifacts are occasionally observed in the output of them. In this work, we address this critical issue by investigating the origins of these artifacts and proposing the conditions to avoid them. First of all, we empirically and theoretically reveal that these problems are caused by "exploding inverse" in the conditional affine coupling layer for certain out-of-distribution (OOD) conditional inputs. Then, we further validated that the probability of causing erroneous artifacts in pixels is highly correlated with a Mahalanobis distance-based OOD score for inverse problems in imaging. Lastly, based on our investigations, we propose a remark to avoid exploding inverse and then based on it, we suggest a simple remedy that substitutes the affine coupling layers with the modified rational quadratic spline coupling layers in normalizing flows, to encourage the robustness of generated image samples. Our experimental results demonstrated that our suggested methods effectively suppressed critical artifacts occurring in normalizing flows for super-resolution space generation and low-light image enhancement.
Open AccessOn the Robustness of Normalizing Flows for Inverse Problems in Imaging
Conditional normalizing flows can generate diverse image samples for solving inverse problems. Most normalizing flows for inverse problems in imaging employ the conditional affine coupling layer that can generate diverse images quickly. However, unintended severe artifacts are occasionally observed in the output of them. In this work, we address this critical issue by investigating the origins of these artifacts and proposing the conditions to avoid them. First of all, we empirically and theoretically reveal that these problems are caused by "exploding inverse" in the conditional affine coupling layer for certain out-of-distribution (OOD) conditional inputs. Then, we further validated that the probability of causing erroneous artifacts in pixels is highly correlated with a Mahalanobis distance-based OOD score for inverse problems in imaging. Lastly, based on our investigations, we propose a remark to avoid exploding inverse and then based on it, we suggest a simple remedy that substitutes the affine coupling layers with the modified rational quadratic spline coupling layers in normalizing flows, to encourage the robustness of generated image samples. Our experimental results demonstrated that our suggested methods effectively suppressed critical artifacts occurring in normalizing flows for super-resolution space generation and low-light image enhancement.
- 18 February 2023
- Published by Springer Nature
- p. 92-129
Efficient and Accurate Quantized Image Super-Resolution on Mobile NPUs, Mobile AI & AIM 2022 Challenge: Report
- 18 February 2023
- Published by Springer Nature
- p. 92-129
Efficient and Accurate Quantized Image Super-Resolution on Mobile NPUs, Mobile AI & AIM 2022 Challenge: Report
- 29 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 21 (22) , 25843-25852
Radar Signal Decomposition in Steering Vector Space for Multi-Target Classification
In automotive frequency-modulated continuous wave radar systems, targets with similar ranges and velocities might be detected as a single target because their target information overlaps on the range-velocity (RV) plane. When the information of multiple targets overlaps on the RV plane, target classification cannot be performed accurately. Therefore, we propose a method to decompose the overlapped target information on the RV plane for effective target classification. To separate each target’s radar signal, we use an orthogonal projection, in which the steering vector formed from the angle information of the desired target is projected onto the kernel space composed of other steering vectors. To verify the performance of the proposed method, a convolutional neural network-based target classification is performed on the decomposed radar signals. Compared to the angle-fast Fourier transform (FFT)-based signal decomposition, the orthogonal projection-based signal decomposition shows higher classification accuracy, which implies that the target information overlapped on the RV plane is effectively separated. When information of two targets is overlapped, the classification accuracy of the proposed method is 5.9%p higher than that of the conventional method. Moreover, when the information of three targets is overlapped, the accuracy is improved by 25.5%p.
- 29 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 21 (22) , 25843-25852
Radar Signal Decomposition in Steering Vector Space for Multi-Target Classification
In automotive frequency-modulated continuous wave radar systems, targets with similar ranges and velocities might be detected as a single target because their target information overlaps on the range-velocity (RV) plane. When the information of multiple targets overlaps on the RV plane, target classification cannot be performed accurately. Therefore, we propose a method to decompose the overlapped target information on the RV plane for effective target classification. To separate each target’s radar signal, we use an orthogonal projection, in which the steering vector formed from the angle information of the desired target is projected onto the kernel space composed of other steering vectors. To verify the performance of the proposed method, a convolutional neural network-based target classification is performed on the decomposed radar signals. Compared to the angle-fast Fourier transform (FFT)-based signal decomposition, the orthogonal projection-based signal decomposition shows higher classification accuracy, which implies that the target information overlapped on the RV plane is effectively separated. When information of two targets is overlapped, the classification accuracy of the proposed method is 5.9%p higher than that of the conventional method. Moreover, when the information of three targets is overlapped, the accuracy is improved by 25.5%p.