Pier Luigi Dragotti
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Pier Luigi Dragotti has a total of 218 co-authors. Here are the top co-authors by number of shared publications:
- Martin Vetterli
School of Computer and Communication Sciences, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland
19 shared publications - Herman Verinaz-Jadan
Faculty of Electrical and Computer Engineering, Escuela Superior Politecnica del Litoral (ESPOL), Guayaquil, Ecuador
13 shared publications - Pingfan Song
Information Engineering, Department of Engineering, University of Cambridge, Cambridge, U.K.
13 shared publications - Amanda Joy Foust
Imperial College London
12 shared publications - Carmel L. Howe
Department of Chemical Physiology and Biochemistry, Oregon Health and Science University, Portland, OR, USA
12 shared publications - Simon R. Schultz
Department of Bioengineering, Imperial College London, London, United Kingdom
12 shared publications
Latest Publications
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 74, 15092-15108
SING: Semantic Image Communications Using Null-Space and INN-Guided Diffusion Models
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 74, 15092-15108
SING: Semantic Image Communications Using Null-Space and INN-Guided Diffusion Models
- Open Access
- 25 November 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122 (48)
Light-field deep learning enables high-throughput, scattering-mitigated calcium imaging
Light-field microscopy (LFM) enables high-throughput functional imaging by scanlessly encoding entire volumes in single snapshots. However, LFM’s computational burden and vulnerability to scattering limit its application to biological imaging. We present a light-field strategy for volumetric, scattering-mitigated neural circuit activity monitoring. A physics-based deep neural network, 2PiLnet, is trained with two-photon volumes and one-photon light fields. Light-field videos of jGCaMP8f-expressing neurons are acquired in neocortical brain slices. 2PiLnet reconstructs volumes with two-photon-like contrast and source confinement from scattered, blurry one-photon light fields from fields-of-view for which no two-photon images are provided. This enables automated segmentation and extraction of calcium signals with high signal-to-noise ratios and reduces optical crosstalk compared to conventional volume reconstruction methods. Imaging 100 volumes per second, we observe putative spikes fired at up to 10 Hz and the spatial intermingling of putative ensembles throughout 530 × 530 × 100 -micron volumes. Compared to iterative algorithms, 2PiLnet workflows reduce light-field video processing times by several-fold, advancing the goal of real-time, scattering-robust volumetric neural circuit imaging for closed-loop and adaptive experimental paradigms.
- 25 November 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122 (48)
Open AccessLight-field deep learning enables high-throughput, scattering-mitigated calcium imaging
Light-field microscopy (LFM) enables high-throughput functional imaging by scanlessly encoding entire volumes in single snapshots. However, LFM’s computational burden and vulnerability to scattering limit its application to biological imaging. We present a light-field strategy for volumetric, scattering-mitigated neural circuit activity monitoring. A physics-based deep neural network, 2PiLnet, is trained with two-photon volumes and one-photon light fields. Light-field videos of jGCaMP8f-expressing neurons are acquired in neocortical brain slices. 2PiLnet reconstructs volumes with two-photon-like contrast and source confinement from scattered, blurry one-photon light fields from fields-of-view for which no two-photon images are provided. This enables automated segmentation and extraction of calcium signals with high signal-to-noise ratios and reduces optical crosstalk compared to conventional volume reconstruction methods. Imaging 100 volumes per second, we observe putative spikes fired at up to 10 Hz and the spatial intermingling of putative ensembles throughout 530 × 530 × 100 -micron volumes. Compared to iterative algorithms, 2PiLnet workflows reduce light-field video processing times by several-fold, advancing the goal of real-time, scattering-robust volumetric neural circuit imaging for closed-loop and adaptive experimental paradigms.
- Open Access
LatentINDIGO: An INN-Guided Latent Diffusion Algorithm for Image Restoration
There is a growing interest in the use of latent diffusion models (LDMs) for image restoration (IR) tasks due to their ability to model effectively the distribution of natural images. While significant progress has been made, there are still key challenges that need to be addressed. First, many approaches depend on a predefined degradation operator, making them ill-suited for complex or unknown degradations that deviate from standard analytical models. Second, many methods struggle to provide a stable guidance in the latent space and finally most methods convert latent representations back to the pixel domain for guidance at every sampling iteration, which significantly increases computational and memory overhead. To overcome these limitations, we introduce a wavelet-inspired invertible neural network (INN) that simulates degradations through a forward transform and reconstructs lost details via the inverse transform. We further integrate this design into a latent diffusion pipeline through two proposed approaches: LatentINDIGO-PixelINN, which operates in the pixel domain, and LatentINDIGO-LatentINN, which stays fully in the latent space to reduce complexity. Both approaches alternate between updating intermediate latent variables under the guidance of our INN and refining the INN forward model to handle unknown degradations. In addition, a regularization step preserves the proximity of latent variables to the natural image manifold. Experiments demonstrate that our algorithm achieves state-of-the-art performance on synthetic and real-world low-quality images, and can be readily adapted to arbitrary output sizes.
Open AccessLatentINDIGO: An INN-Guided Latent Diffusion Algorithm for Image Restoration
There is a growing interest in the use of latent diffusion models (LDMs) for image restoration (IR) tasks due to their ability to model effectively the distribution of natural images. While significant progress has been made, there are still key challenges that need to be addressed. First, many approaches depend on a predefined degradation operator, making them ill-suited for complex or unknown degradations that deviate from standard analytical models. Second, many methods struggle to provide a stable guidance in the latent space and finally most methods convert latent representations back to the pixel domain for guidance at every sampling iteration, which significantly increases computational and memory overhead. To overcome these limitations, we introduce a wavelet-inspired invertible neural network (INN) that simulates degradations through a forward transform and reconstructs lost details via the inverse transform. We further integrate this design into a latent diffusion pipeline through two proposed approaches: LatentINDIGO-PixelINN, which operates in the pixel domain, and LatentINDIGO-LatentINN, which stays fully in the latent space to reduce complexity. Both approaches alternate between updating intermediate latent variables under the guidance of our INN and refining the INN forward model to handle unknown degradations. In addition, a regularization step preserves the proximity of latent variables to the natural image manifold. Experiments demonstrate that our algorithm achieves state-of-the-art performance on synthetic and real-world low-quality images, and can be readily adapted to arbitrary output sizes.
- Open Access
Light-field deep learning enables high-throughput, scattering-mitigated calcium imaging
Light field microscopy (LFM) enables volumetric, high throughput functional imaging. However, the computational burden and vulnerability to scattering limit LFM’s application to neuroscience. We present a light-field strategy for volumetric, scattering-mitigated neural circuit activity monitoring. A physics-based deep neural network, LNet, is trained with two-photon volumes and one-photon light fields. A processing pipeline uses LNet to extract calcium activity from light-field videos of jGCaMP8f-expressing neurons in acute cortical slices. The extracted time series have high signal-to-noise ratios and reduced optical crosstalk compared to conventional volume reconstruction methods. Imaging 100 volumes per second, we observe putative spikes fired at up to 10 Hz and the spatial intermingling of putative ensembles throughout 530 x 530 x 100-micron volumes. Compared to iterative algorithms, LNet workflows reduce light-field video processing times by 2- to 12-fold, advancing the goal of real-time, scattering-robust volumetric neural circuit imaging for closed-loop and adaptive experimental paradigms.
Open AccessLight-field deep learning enables high-throughput, scattering-mitigated calcium imaging
Light field microscopy (LFM) enables volumetric, high throughput functional imaging. However, the computational burden and vulnerability to scattering limit LFM’s application to neuroscience. We present a light-field strategy for volumetric, scattering-mitigated neural circuit activity monitoring. A physics-based deep neural network, LNet, is trained with two-photon volumes and one-photon light fields. A processing pipeline uses LNet to extract calcium activity from light-field videos of jGCaMP8f-expressing neurons in acute cortical slices. The extracted time series have high signal-to-noise ratios and reduced optical crosstalk compared to conventional volume reconstruction methods. Imaging 100 volumes per second, we observe putative spikes fired at up to 10 Hz and the spatial intermingling of putative ensembles throughout 530 x 530 x 100-micron volumes. Compared to iterative algorithms, LNet workflows reduce light-field video processing times by 2- to 12-fold, advancing the goal of real-time, scattering-robust volumetric neural circuit imaging for closed-loop and adaptive experimental paradigms.
- Open Access
SMILENet: Unleashing Extra-Large Capacity Image Steganography via a Synergistic Mosaic InvertibLE Hiding Network
Existing image steganography methods face fundamental limitations in hiding capacity (typically $1\sim7$ images) due to severe information interference and uncoordinated capacity-distortion trade-off. We propose SMILENet, a novel synergistic framework that achieves 25 image hiding through three key innovations: (i) A synergistic network architecture coordinates reversible and non-reversible operations to efficiently exploit information redundancy in both secret and cover images. The reversible Invertible Cover-Driven Mosaic (ICDM) module and Invertible Mosaic Secret Embedding (IMSE) module establish cover-guided mosaic transformations and representation embedding with mathematically guaranteed invertibility for distortion-free embedding. The non-reversible Secret Information Selection (SIS) module and Secret Detail Enhancement (SDE) module implement learnable feature modulation for critical information selection and enhancement. (ii) A unified training strategy that coordinates complementary modules to achieve 3.0x higher capacity than existing methods with superior visual quality. (iii) Last but not least, we introduce a new metric to model Capacity-Distortion Trade-off for evaluating the image steganography algorithms that jointly considers hiding capacity and distortion, and provides a unified evaluation approach for accessing results with different number of secret image. Extensive experiments on DIV2K, Paris StreetView and ImageNet1K show that SMILENet outperforms state-of-the-art methods in terms of hiding capacity, recovery quality as well as security against steganalysis methods.Open AccessSMILENet: Unleashing Extra-Large Capacity Image Steganography via a Synergistic Mosaic InvertibLE Hiding Network
Existing image steganography methods face fundamental limitations in hiding capacity (typically $1\sim7$ images) due to severe information interference and uncoordinated capacity-distortion trade-off. We propose SMILENet, a novel synergistic framework that achieves 25 image hiding through three key innovations: (i) A synergistic network architecture coordinates reversible and non-reversible operations to efficiently exploit information redundancy in both secret and cover images. The reversible Invertible Cover-Driven Mosaic (ICDM) module and Invertible Mosaic Secret Embedding (IMSE) module establish cover-guided mosaic transformations and representation embedding with mathematically guaranteed invertibility for distortion-free embedding. The non-reversible Secret Information Selection (SIS) module and Secret Detail Enhancement (SDE) module implement learnable feature modulation for critical information selection and enhancement. (ii) A unified training strategy that coordinates complementary modules to achieve 3.0x higher capacity than existing methods with superior visual quality. (iii) Last but not least, we introduce a new metric to model Capacity-Distortion Trade-off for evaluating the image steganography algorithms that jointly considers hiding capacity and distortion, and provides a unified evaluation approach for accessing results with different number of secret image. Extensive experiments on DIV2K, Paris StreetView and ImageNet1K show that SMILENet outperforms state-of-the-art methods in terms of hiding capacity, recovery quality as well as security against steganalysis methods. - 6 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 47 (6) , 4957-4973
A Lightweight Deep Exclusion Unfolding Network for Single Image Reflection Removal
Single Image Reflection Removal (SIRR) is a canonical blind source separation problem and refers to the issue of separating a reflection-contaminated image into a transmission and a reflection image. The core challenge lies in minimizing the commonalities among different sources. Existing deep learning approaches either neglect the significance of feature interactions or rely on heuristically designed architectures. In this paper, we propose a novel Deep Exclusion unfolding Network (DExNet), a lightweight, interpretable, and effective network architecture for SIRR. DExNet is principally constructed by unfolding and parameterizing a simple iterative Sparse and Auxiliary Feature Update (i-SAFU) algorithm, which is specifically designed to solve a new model-based SIRR optimization formulation incorporating a general exclusion prior. This general exclusion prior enables the unfolded SAFU module to inherently identify and penalize commonalities between the transmission and reflection features, ensuring more accurate separation. The principled design of DExNet not only enhances its interpretability but also significantly improves its performance. Comprehensive experiments on four benchmark datasets demonstrate that DExNet achieves state-of-the-art visual and quantitative results while utilizing only approximately 8% of the parameters required by leading methods.
- 6 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 47 (6) , 4957-4973
A Lightweight Deep Exclusion Unfolding Network for Single Image Reflection Removal
Single Image Reflection Removal (SIRR) is a canonical blind source separation problem and refers to the issue of separating a reflection-contaminated image into a transmission and a reflection image. The core challenge lies in minimizing the commonalities among different sources. Existing deep learning approaches either neglect the significance of feature interactions or rely on heuristically designed architectures. In this paper, we propose a novel Deep Exclusion unfolding Network (DExNet), a lightweight, interpretable, and effective network architecture for SIRR. DExNet is principally constructed by unfolding and parameterizing a simple iterative Sparse and Auxiliary Feature Update (i-SAFU) algorithm, which is specifically designed to solve a new model-based SIRR optimization formulation incorporating a general exclusion prior. This general exclusion prior enables the unfolded SAFU module to inherently identify and penalize commonalities between the transmission and reflection features, ensuring more accurate separation. The principled design of DExNet not only enhances its interpretability but also significantly improves its performance. Comprehensive experiments on four benchmark datasets demonstrate that DExNet achieves state-of-the-art visual and quantitative results while utilizing only approximately 8% of the parameters required by leading methods.
- Open Access
A Lightweight Deep Exclusion Unfolding Network for Single Image Reflection Removal
Single Image Reflection Removal (SIRR) is a canonical blind source separation problem and refers to the issue of separating a reflection-contaminated image into a transmission and a reflection image. The core challenge lies in minimizing the commonalities among different sources. Existing deep learning approaches either neglect the significance of feature interactions or rely on heuristically designed architectures. In this paper, we propose a novel Deep Exclusion unfolding Network (DExNet), a lightweight, interpretable, and effective network architecture for SIRR. DExNet is principally constructed by unfolding and parameterizing a simple iterative Sparse and Auxiliary Feature Update (i-SAFU) algorithm, which is specifically designed to solve a new model-based SIRR optimization formulation incorporating a general exclusion prior. This general exclusion prior enables the unfolded SAFU module to inherently identify and penalize commonalities between the transmission and reflection features, ensuring more accurate separation. The principled design of DExNet not only enhances its interpretability but also significantly improves its performance. Comprehensive experiments on four benchmark datasets demonstrate that DExNet achieves state-of-the-art visual and quantitative results while utilizing only approximately 8\% of the parameters required by leading methods.
Open AccessA Lightweight Deep Exclusion Unfolding Network for Single Image Reflection Removal
Single Image Reflection Removal (SIRR) is a canonical blind source separation problem and refers to the issue of separating a reflection-contaminated image into a transmission and a reflection image. The core challenge lies in minimizing the commonalities among different sources. Existing deep learning approaches either neglect the significance of feature interactions or rely on heuristically designed architectures. In this paper, we propose a novel Deep Exclusion unfolding Network (DExNet), a lightweight, interpretable, and effective network architecture for SIRR. DExNet is principally constructed by unfolding and parameterizing a simple iterative Sparse and Auxiliary Feature Update (i-SAFU) algorithm, which is specifically designed to solve a new model-based SIRR optimization formulation incorporating a general exclusion prior. This general exclusion prior enables the unfolded SAFU module to inherently identify and penalize commonalities between the transmission and reflection features, ensuring more accurate separation. The principled design of DExNet not only enhances its interpretability but also significantly improves its performance. Comprehensive experiments on four benchmark datasets demonstrate that DExNet achieves state-of-the-art visual and quantitative results while utilizing only approximately 8\% of the parameters required by leading methods.
- Open Access
- 16 December 2024
- Published by Elsevier in Neural Networks
- Vol. 184, 107043
Enhanced accuracy in first-spike coding using current-based adaptive LIF neuron
- 16 December 2024
- Published by Elsevier in Neural Networks
- Vol. 184, 107043
Open AccessEnhanced accuracy in first-spike coding using current-based adaptive LIF neuron
- Open Access
Adversarial Deep-Unfolding Network for MA-XRF Super-Resolution on Old Master Paintings Using Minimal Training Data
High-quality element distribution maps enable precise analysis of the material composition and condition of Old Master paintings. These maps are typically produced from data acquired through Macro X-ray fluorescence (MA-XRF) scanning, a non-invasive technique that collects spectral information. However, MA-XRF is often limited by a trade-off between acquisition time and resolution. Achieving higher resolution requires longer scanning times, which can be impractical for detailed analysis of large artworks. Super-resolution MA-XRF provides an alternative solution by enhancing the quality of MA-XRF scans while reducing the need for extended scanning sessions. This paper introduces a tailored super-resolution approach to improve MA-XRF analysis of Old Master paintings. Our method proposes a novel adversarial neural network architecture for MA-XRF, inspired by the Learned Iterative Shrinkage-Thresholding Algorithm. It is specifically designed to work in an unsupervised manner, making efficient use of the limited available data. This design avoids the need for extensive datasets or pre-trained networks, allowing it to be trained using just a single high-resolution RGB image alongside low-resolution MA-XRF data. Numerical results demonstrate that our method outperforms existing state-of-the-art super-resolution techniques for MA-XRF scans of Old Master paintings.
Open AccessAdversarial Deep-Unfolding Network for MA-XRF Super-Resolution on Old Master Paintings Using Minimal Training Data
High-quality element distribution maps enable precise analysis of the material composition and condition of Old Master paintings. These maps are typically produced from data acquired through Macro X-ray fluorescence (MA-XRF) scanning, a non-invasive technique that collects spectral information. However, MA-XRF is often limited by a trade-off between acquisition time and resolution. Achieving higher resolution requires longer scanning times, which can be impractical for detailed analysis of large artworks. Super-resolution MA-XRF provides an alternative solution by enhancing the quality of MA-XRF scans while reducing the need for extended scanning sessions. This paper introduces a tailored super-resolution approach to improve MA-XRF analysis of Old Master paintings. Our method proposes a novel adversarial neural network architecture for MA-XRF, inspired by the Learned Iterative Shrinkage-Thresholding Algorithm. It is specifically designed to work in an unsupervised manner, making efficient use of the limited available data. This design avoids the need for extensive datasets or pre-trained networks, allowing it to be trained using just a single high-resolution RGB image alongside low-resolution MA-XRF data. Numerical results demonstrate that our method outperforms existing state-of-the-art super-resolution techniques for MA-XRF scans of Old Master paintings.
- Open Access
- 9 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 18 (6) , 1108-1122
INDIGO+: A Unified INN-Guided Probabilistic Diffusion Algorithm for Blind and Non-Blind Image Restoration
Generative diffusion models are becoming one of the most popular prior in image restoration (IR) tasks due to their remarkable ability to generate realistic natural images. Despite achieving satisfactory results, IR methods based on diffusion models present several limitations. First of all, most non-blind approaches require an analytical expression of the degradation model to guide the sampling process. Secondly, most existing blind approaches rely on families of pre-defined degradation models for training their deep networks. The above issues limit the flexibility of these approaches and so their ability to handle real-world degradation tasks. In this paper, we propose a novel INN-guided probabilistic diffusion algorithm for non-blind and blind image restoration, namely INDIGO and BlindINDIGO, which combines the merits of the perfect reconstruction property of invertible neural networks (INN) with the strong generative capabilities of pre-trained diffusion models. Specifically, we train the forward process of the INN to simulate an arbitrary degradation process and use the inverse to obtain an intermediate image that we use to guide the reverse diffusion sampling process through a gradient step. We also introduce an initialization strategy, to further improve the performance and inference speed of our algorithm. Experiments demonstrate that our algorithm obtains competitive results compared with recently leading methods both quantitatively and visually on synthetic and real-world low-quality images.
- 9 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 18 (6) , 1108-1122
Open AccessINDIGO+: A Unified INN-Guided Probabilistic Diffusion Algorithm for Blind and Non-Blind Image Restoration
Generative diffusion models are becoming one of the most popular prior in image restoration (IR) tasks due to their remarkable ability to generate realistic natural images. Despite achieving satisfactory results, IR methods based on diffusion models present several limitations. First of all, most non-blind approaches require an analytical expression of the degradation model to guide the sampling process. Secondly, most existing blind approaches rely on families of pre-defined degradation models for training their deep networks. The above issues limit the flexibility of these approaches and so their ability to handle real-world degradation tasks. In this paper, we propose a novel INN-guided probabilistic diffusion algorithm for non-blind and blind image restoration, namely INDIGO and BlindINDIGO, which combines the merits of the perfect reconstruction property of invertible neural networks (INN) with the strong generative capabilities of pre-trained diffusion models. Specifically, we train the forward process of the INN to simulate an arbitrary degradation process and use the inverse to obtain an intermediate image that we use to guide the reverse diffusion sampling process through a gradient step. We also introduce an initialization strategy, to further improve the performance and inference speed of our algorithm. Experiments demonstrate that our algorithm obtains competitive results compared with recently leading methods both quantitatively and visually on synthetic and real-world low-quality images.