Laurent D. Cohen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Laurent D. Cohen has a total of 69 co-authors. Here are the top co-authors by number of shared publications:
- Da Chen
Shandong Artificial Intelligence Institute, Qilu University of Technology (Shandong Academy of Sciences), Jinan, China
17 shared publications - Gabriel Peyré
And ENS CNRS PSL Université
7 shared publications - Minglei Shu
Shandong Artificial Intelligence Institute, Qilu University of Technology (Shandong Academy of Sciences), Jinan, China
7 shared publications - Jean-Marie Mirebeau
Department of Mathematics, Centre Borelli, ENS Paris-Saclay, CNRS, University Paris-Saclay, Gif-sur-Yvette, France
6 shared publications - Fethallah Benmansour
Ceremade, UMR CNRS 7534, Université Paris-Dauphine, Paris Cedex 16, France
3 shared publications - Li Liu
Yuanshen Rehabilitation Institute, Shanghai Jiao Tong University School of Medicine, Shanghai, China
3 shared publications
Latest Publications
- Open Access
- 1 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 47 (12) , 11107-11123
Neovascularization Segmentation via a Multilateral Interaction-Enhanced Graph Convolutional Network
Choroidal neovascularization (CNV), a primary characteristic of wet age-related macular degeneration (wet AMD), represents a leading cause of blindness worldwide. In clinical practice, optical coherence tomography angiography (OCTA) is commonly used for studying CNV-related pathological changes, due to its micron-level resolution and non-invasive nature. Thus, accurate segmentation of CNV regions and vessels in OCTA images is crucial for clinical assessment of wet AMD. However, challenges existed due to irregular CNV shapes and imaging limitations like projection artifacts, noises and boundary blurring. Moreover, the lack of publicly available datasets constraints the CNV analysis. To address these challenges, this paper constructs the first publicly accessible CNV dataset (CNVSeg), and proposes a novel multilateral graph convolutional interaction-enhanced CNV segmentation network (MTG-Net). This network integrates both region and vessel morphological information, exploring semantic and geometric duality constraints within the graph domain. Specifically, MTG-Net consists of a multi-task framework and two graph-based cross-task modules: Multilateral Interaction Graph Reasoning (MIGR) and Multilateral Reinforcement Graph Reasoning (MRGR). The multi-task framework encodes rich geometric features of lesion shapes and surfaces, decoupling the image into three task-specific feature maps. MIGR and MRGR iteratively reason about higher-order relationships across tasks through a graph mechanism, enabling complementary optimization for task-specific objectives. Additionally, an uncertainty-weighted loss is proposed to mitigate the impact of artifacts and noise on segmentation accuracy. Experimental results demonstrate that MTG-Net outperforms existing methods, achieving a Dice socre of 87.21% for region segmentation and 88.12% for vessel segmentation.
- 1 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 47 (12) , 11107-11123
Open AccessNeovascularization Segmentation via a Multilateral Interaction-Enhanced Graph Convolutional Network
Choroidal neovascularization (CNV), a primary characteristic of wet age-related macular degeneration (wet AMD), represents a leading cause of blindness worldwide. In clinical practice, optical coherence tomography angiography (OCTA) is commonly used for studying CNV-related pathological changes, due to its micron-level resolution and non-invasive nature. Thus, accurate segmentation of CNV regions and vessels in OCTA images is crucial for clinical assessment of wet AMD. However, challenges existed due to irregular CNV shapes and imaging limitations like projection artifacts, noises and boundary blurring. Moreover, the lack of publicly available datasets constraints the CNV analysis. To address these challenges, this paper constructs the first publicly accessible CNV dataset (CNVSeg), and proposes a novel multilateral graph convolutional interaction-enhanced CNV segmentation network (MTG-Net). This network integrates both region and vessel morphological information, exploring semantic and geometric duality constraints within the graph domain. Specifically, MTG-Net consists of a multi-task framework and two graph-based cross-task modules: Multilateral Interaction Graph Reasoning (MIGR) and Multilateral Reinforcement Graph Reasoning (MRGR). The multi-task framework encodes rich geometric features of lesion shapes and surfaces, decoupling the image into three task-specific feature maps. MIGR and MRGR iteratively reason about higher-order relationships across tasks through a graph mechanism, enabling complementary optimization for task-specific objectives. Additionally, an uncertainty-weighted loss is proposed to mitigate the impact of artifacts and noise on segmentation accuracy. Experimental results demonstrate that MTG-Net outperforms existing methods, achieving a Dice socre of 87.21% for region segmentation and 88.12% for vessel segmentation.
- 17 May 2025
- Published by Springer Nature
- p. 308-321
Learning Anisotropic Metrics for Geodesic Distances via the Heat Equation for Image Segmentation
- 17 May 2025
- Published by Springer Nature
- p. 308-321
Learning Anisotropic Metrics for Geodesic Distances via the Heat Equation for Image Segmentation
- 29 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-14
DPSNet: A Detail Perception Synergistic Network for Camouflaged Object Detection
Camouflaged object detection (COD) is a challenging task that aims to identify objects that blend into their backgrounds. This has various applications in computer vision, including military surveillance, animal behavior analysis, and image editing. However, existing COD methods often face issues of ambiguous boundaries and inaccurate localization due to ineffective feature fusion and boundary guidance. To address these challenges, we propose a new hierarchical synergistic prediction framework, named Detail Perception Synergistic Network (DPSNet), which leverages an incremental strategy and a novel detail perception loss for accurate COD. Specifically, our DPSNet consists of a baseline network and an enhanced network, generating preliminary and fine-grained predictions, respectively. To adapt to the morphological variations of camouflaged objects, we design a Morphological Awareness Adaptation Module (MAAM) that adaptively fuses inconsistent morphological and texture features. To enhance the contextual semantic feature representations and obtain clear predictions, we propose a Semantic Aggregation Enhancement Module (SAEM) that aggregates contextual semantic features with a deep aggregate group with different receptive fields. Furthermore, to guide the network in learning boundary information for more accurate predictions, we present a Detail Perception Loss (DP Loss) that balances global information and hard samples. Without any additional tricks, extensive experiments demonstrate that our proposed DPSNet achieves real-time performance and significantly surpasses 37 state-of-the-art methods across four standard metrics on four widely used benchmark datasets. The code will be available at https://github.com/fxle/DPSNet.
- 29 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-14
DPSNet: A Detail Perception Synergistic Network for Camouflaged Object Detection
Camouflaged object detection (COD) is a challenging task that aims to identify objects that blend into their backgrounds. This has various applications in computer vision, including military surveillance, animal behavior analysis, and image editing. However, existing COD methods often face issues of ambiguous boundaries and inaccurate localization due to ineffective feature fusion and boundary guidance. To address these challenges, we propose a new hierarchical synergistic prediction framework, named Detail Perception Synergistic Network (DPSNet), which leverages an incremental strategy and a novel detail perception loss for accurate COD. Specifically, our DPSNet consists of a baseline network and an enhanced network, generating preliminary and fine-grained predictions, respectively. To adapt to the morphological variations of camouflaged objects, we design a Morphological Awareness Adaptation Module (MAAM) that adaptively fuses inconsistent morphological and texture features. To enhance the contextual semantic feature representations and obtain clear predictions, we propose a Semantic Aggregation Enhancement Module (SAEM) that aggregates contextual semantic features with a deep aggregate group with different receptive fields. Furthermore, to guide the network in learning boundary information for more accurate predictions, we present a Detail Perception Loss (DP Loss) that balances global information and hard samples. Without any additional tricks, extensive experiments demonstrate that our proposed DPSNet achieves real-time performance and significantly surpasses 37 state-of-the-art methods across four standard metrics on four widely used benchmark datasets. The code will be available at https://github.com/fxle/DPSNet.
- 20 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-14
A Generalized Geodesic Voting Framework for Interactive Image Segmentation
In this article, we introduce a new variational model for addressing the image segmentation problem of minimal user interaction. The proposed variational segmentation model, regarded as a variant of the classical geodesic voting method, is designed to take advantage of the topology constraint of object boundary contour and the local homogeneity of image gray levels. In particular, the proposed model takes into account a metric of the asymmetric quadratic form to build the path energy of the boundary contour, allowing it to encode the local image intensity consistency for computing minimal paths. Furthermore, an adaptive cut-based closed contour computation scheme is invoked to depict the target boundary contour, implemented by computing two minimal paths traveling along two different directions from a single source point located at the adaptive cut. Both paths form a simple closed curve that links the two sides of the cut. Finally, experimental results show that the proposed geodesic voting model can deal with complicated segmentation tasks and indeed outperforms state-of-the-art image segmentation approaches, especially for medical image segmentation problems.
- 20 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-14
A Generalized Geodesic Voting Framework for Interactive Image Segmentation
In this article, we introduce a new variational model for addressing the image segmentation problem of minimal user interaction. The proposed variational segmentation model, regarded as a variant of the classical geodesic voting method, is designed to take advantage of the topology constraint of object boundary contour and the local homogeneity of image gray levels. In particular, the proposed model takes into account a metric of the asymmetric quadratic form to build the path energy of the boundary contour, allowing it to encode the local image intensity consistency for computing minimal paths. Furthermore, an adaptive cut-based closed contour computation scheme is invoked to depict the target boundary contour, implemented by computing two minimal paths traveling along two different directions from a single source point located at the adaptive cut. Both paths form a simple closed curve that links the two sides of the cut. Finally, experimental results show that the proposed geodesic voting model can deal with complicated segmentation tasks and indeed outperforms state-of-the-art image segmentation approaches, especially for medical image segmentation problems.
- 22 August 2024
- Published by Springer Nature
- p. 24-47
Unified Shape Analysis and Synthesis via Deformable Voxel Grids
- 22 August 2024
- Published by Springer Nature
- p. 24-47
Unified Shape Analysis and Synthesis via Deformable Voxel Grids
- 12 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 33 (10577149) , 793-808
Grouping Boundary Proposals for Fast Interactive Image Segmentation
Geodesic models are known as an efficient tool for solving various image segmentation problems. Most of existing approaches only exploit local pointwise image features to track geodesic paths for delineating the objective boundaries. However, such a segmentation strategy cannot take into account the connectivity of the image edge features, increasing the risk of shortcut problem, especially in the case of complicated scenario. In this work, we introduce a new image segmentation model based on the minimal geodesic framework in conjunction with an adaptive cut-based circular optimal path computation scheme and a graph-based boundary proposals grouping scheme. Specifically, the adaptive cut can disconnect the image domain such that the target contours are imposed to pass through this cut only once. The boundary proposals are comprised of precomputed image edge segments, providing the connectivity information for our segmentation model. These boundary proposals are then incorporated into the proposed image segmentation model, such that the target segmentation contours are made up of a set of selected boundary proposals and the corresponding geodesic paths linking them. Experimental results show that the proposed model indeed outperforms state-of-the-art minimal paths-based image segmentation approaches.
- 12 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 33 (10577149) , 793-808
Grouping Boundary Proposals for Fast Interactive Image Segmentation
Geodesic models are known as an efficient tool for solving various image segmentation problems. Most of existing approaches only exploit local pointwise image features to track geodesic paths for delineating the objective boundaries. However, such a segmentation strategy cannot take into account the connectivity of the image edge features, increasing the risk of shortcut problem, especially in the case of complicated scenario. In this work, we introduce a new image segmentation model based on the minimal geodesic framework in conjunction with an adaptive cut-based circular optimal path computation scheme and a graph-based boundary proposals grouping scheme. Specifically, the adaptive cut can disconnect the image domain such that the target contours are imposed to pass through this cut only once. The boundary proposals are comprised of precomputed image edge segments, providing the connectivity information for our segmentation model. These boundary proposals are then incorporated into the proposed image segmentation model, such that the target segmentation contours are made up of a set of selected boundary proposals and the corresponding geodesic paths linking them. Experimental results show that the proposed model indeed outperforms state-of-the-art minimal paths-based image segmentation approaches.
- Open Access
- 24 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 178-185
DeepPrism: Channel Convolution for Lightweight Generative Models
In this paper, we introduce DeepPrism, a novel network architecture for generative models, which addresses the redundancy issue of convolutional filters. DeepPrism reaches an unprecedented parameter efficiency improvement of up to 1000 times on generative models. The number of parameters is reduced over conventional CNNs from quadratic to constant dependency with respect to the network's width. The training / inference time and space are also reduced. The main novelty lies in the geometric property namely the translation equivariance on channels, which gives rise to the convolution structure along the channels, and trivially generalizes to the attention mechanism. Compared with Latent Diffusion Model, DeepPrism produces similar qualitative and quantitative results, but the number of parameters can be reduced at a great scale. Other generative models including autoencoders and single-image diffusion models are also experimented to exhibit the generality of this method on generative models.
- 24 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 178-185
Open AccessDeepPrism: Channel Convolution for Lightweight Generative Models
In this paper, we introduce DeepPrism, a novel network architecture for generative models, which addresses the redundancy issue of convolutional filters. DeepPrism reaches an unprecedented parameter efficiency improvement of up to 1000 times on generative models. The number of parameters is reduced over conventional CNNs from quadratic to constant dependency with respect to the network's width. The training / inference time and space are also reduced. The main novelty lies in the geometric property namely the translation equivariance on channels, which gives rise to the convolution structure along the channels, and trivially generalizes to the attention mechanism. Compared with Latent Diffusion Model, DeepPrism produces similar qualitative and quantitative results, but the number of parameters can be reduced at a great scale. Other generative models including autoencoders and single-image diffusion models are also experimented to exhibit the generality of this method on generative models.
- Open Access
Fitting tree model with CNN and geodesics to track vesselsand application to Ultrasound Localization Microscopy data
Segmentation of tubular structures in vascular imaging is a well studied task, although it is rare that we try to infuse knowledge of the tree-like structure of the regions to be detected. Our work focuses on detecting the important landmarks in the vascular network (via CNN performing both localization and classification of the points of interest) and representing vessels as the edges in some minimal distance tree graph. We leverage geodesic methods relevant to the detection of vessels and their geometry, making use of the space of positions and orientations so that 2D vessels can be accurately represented as trees. We build our model to carry tracking on Ultrasound Localization Microscopy (ULM) data, proposing to build a good cost function for tracking on this type of data. We also test our framework on synthetic and eye fundus data. Results show that scarcity of well annotated ULM data is an obstacle to localization of vascular landmarks but the Orientation Score built from ULM data yields good geodesics for tracking blood vessels.
Open AccessFitting tree model with CNN and geodesics to track vesselsand application to Ultrasound Localization Microscopy data
Segmentation of tubular structures in vascular imaging is a well studied task, although it is rare that we try to infuse knowledge of the tree-like structure of the regions to be detected. Our work focuses on detecting the important landmarks in the vascular network (via CNN performing both localization and classification of the points of interest) and representing vessels as the edges in some minimal distance tree graph. We leverage geodesic methods relevant to the detection of vessels and their geometry, making use of the space of positions and orientations so that 2D vessels can be accurately represented as trees. We build our model to carry tracking on Ultrasound Localization Microscopy (ULM) data, proposing to build a good cost function for tracking on this type of data. We also test our framework on synthetic and eye fundus data. Results show that scarcity of well annotated ULM data is an obstacle to localization of vascular landmarks but the Orientation Score built from ULM data yields good geodesics for tracking blood vessels.
- Open Access
A Region-based Randers Geodesic Approach for Image Segmentation
The geodesic model based on the eikonal partial differential equation (PDE) has served as a fundamental tool for the applications of image segmentation and boundary detection in the past two decades. However, the existing approaches commonly only exploit the image edge-based features for computing minimal geodesic paths, potentially limiting their performance in complicated segmentation situations. In this paper, we introduce a new variational image segmentation model based on the minimal geodesic path framework and the eikonal PDE, where the region-based appearance term that defines then regional homogeneity features can be taken into account for estimating the associated minimal geodesic paths. This is done by constructing a Randers geodesic metric interpretation of the region-based active contour energy functional. As a result, the minimization of the active contour energy functional is transformed into finding the solution to the Randers eikonal PDE. We also suggest a practical interactive image segmentation strategy, where the target boundary can be delineated by the concatenation of several piecewise geodesic paths. We invoke the Finsler variant of the fast marching method to estimate the geodesic distance map, yielding an efficient implementation of the proposed region-based Randers geodesic model for image segmentation. Experimental results on both synthetic and real images exhibit that our model indeed achieves encouraging segmentation performance.
Open AccessA Region-based Randers Geodesic Approach for Image Segmentation
The geodesic model based on the eikonal partial differential equation (PDE) has served as a fundamental tool for the applications of image segmentation and boundary detection in the past two decades. However, the existing approaches commonly only exploit the image edge-based features for computing minimal geodesic paths, potentially limiting their performance in complicated segmentation situations. In this paper, we introduce a new variational image segmentation model based on the minimal geodesic path framework and the eikonal PDE, where the region-based appearance term that defines then regional homogeneity features can be taken into account for estimating the associated minimal geodesic paths. This is done by constructing a Randers geodesic metric interpretation of the region-based active contour energy functional. As a result, the minimization of the active contour energy functional is transformed into finding the solution to the Randers eikonal PDE. We also suggest a practical interactive image segmentation strategy, where the target boundary can be delineated by the concatenation of several piecewise geodesic paths. We invoke the Finsler variant of the fast marching method to estimate the geodesic distance map, yielding an efficient implementation of the proposed region-based Randers geodesic model for image segmentation. Experimental results on both synthetic and real images exhibit that our model indeed achieves encouraging segmentation performance.
- Open Access
Computing Geodesic Paths Encoding a Curvature Prior
In this paper, we introduce an efficient method for computing curves minimizing a variant of the Euler-Mumford elastica energy, with fixed endpoints and tangents at these endpoints, where the bending energy is enhanced with a user defined and data-driven scalar-valued term referred to as the curvature prior. In order to guarantee that the globally optimal curve is extracted, the proposed method involves the numerical computation of the viscosity solution to a specific static Hamilton-Jacobi-Bellman (HJB) partial differential equation (PDE). For that purpose, we derive the explicit Hamiltonian associated to this variant model equipped with a curvature prior, discretize the resulting HJB PDE using an adaptive finite difference scheme, and solve it in a single pass using a generalized Fast-Marching method. In addition, we also present a practical method for estimating the curvature prior values from image data, designed for the task of accurately tracking curvilinear structure centerlines. Numerical experiments on synthetic and real image data illustrate the advantages of the considered variant of the elastica model with a prior curvature enhancement in complex scenarios where challenging geometric structures appear.
Open AccessComputing Geodesic Paths Encoding a Curvature Prior
In this paper, we introduce an efficient method for computing curves minimizing a variant of the Euler-Mumford elastica energy, with fixed endpoints and tangents at these endpoints, where the bending energy is enhanced with a user defined and data-driven scalar-valued term referred to as the curvature prior. In order to guarantee that the globally optimal curve is extracted, the proposed method involves the numerical computation of the viscosity solution to a specific static Hamilton-Jacobi-Bellman (HJB) partial differential equation (PDE). For that purpose, we derive the explicit Hamiltonian associated to this variant model equipped with a curvature prior, discretize the resulting HJB PDE using an adaptive finite difference scheme, and solve it in a single pass using a generalized Fast-Marching method. In addition, we also present a practical method for estimating the curvature prior values from image data, designed for the task of accurately tracking curvilinear structure centerlines. Numerical experiments on synthetic and real image data illustrate the advantages of the considered variant of the elastica model with a prior curvature enhancement in complex scenarios where challenging geometric structures appear.