Morteza Ghahremani
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
Morteza Ghahremani has a total of 36 co-authors. Here are the top co-authors by number of shared publications:
- Bernard Tiddeman
Department of Computer Science, Aberystwyth University, Aberystwyth, United Kingdom
4 shared publications - Hassan Ghassemian
Image Processing and Information Analysis Lab, Faculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, Iran
4 shared publications - Yonghuai Liu
Department of Computer Science, Edge Hill University, Ormskirk, U.K.
4 shared publications - Bailiang Jian
Technical University of Munich
3 shared publications - Christian Wachinger
Technical University of Munich
3 shared publications - Jiazhen Pan
Technical University of Munich
3 shared publications
Latest Publications
- 2 July 2026
- Published by Springer Nature
- p. 43-51
Unleashing the Power of Intensity Augmentation for Multi-modal Image Registration
- 2 July 2026
- Published by Springer Nature
- p. 43-51
Unleashing the Power of Intensity Augmentation for Multi-modal Image Registration
- 9 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 27 (15209210) , 8393-8407
Box It to Bind It: Unified Layout Control and Attribute Binding in Text-to-Image Diffusion Models
While latent diffusion models (LDMs) excel at creating imaginative images, they often lack precision in semantic fidelity and spatial control over where objects are generated. To address these deficiencies, we introduce the Box-it-to-Bind-it (B2B) module-a novel, training-free approach for improving spatial control and semantic accuracy in text-to-image (T2I) diffusion models. B2B targets three key challenges in T2I: catastrophic neglect, attribute binding, and layout guidance. The process encompasses two main steps: (i) Object generation, which adjusts the latent encoding to guarantee object generation and directs it within specified bounding boxes, and (ii) Attribute binding, ensuring that generated objects adhere to their specified attributes in the prompt. B2B is designed as a compatible plug-and-play module for existing T2I models like Stable Diffusion and Gligen, markedly enhancing models' performance in addressing these key challenges. We assess our technique on the well-established CompBench and TIFA score benchmarks, and HRS dataset where B2B not only surpasses methods specialized in either attribute binding or layout guidance but also uniquely excels by integrating these capabilities to deliver enhanced overall performance.
- 9 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 27 (15209210) , 8393-8407
Box It to Bind It: Unified Layout Control and Attribute Binding in Text-to-Image Diffusion Models
While latent diffusion models (LDMs) excel at creating imaginative images, they often lack precision in semantic fidelity and spatial control over where objects are generated. To address these deficiencies, we introduce the Box-it-to-Bind-it (B2B) module-a novel, training-free approach for improving spatial control and semantic accuracy in text-to-image (T2I) diffusion models. B2B targets three key challenges in T2I: catastrophic neglect, attribute binding, and layout guidance. The process encompasses two main steps: (i) Object generation, which adjusts the latent encoding to guarantee object generation and directs it within specified bounding boxes, and (ii) Attribute binding, ensuring that generated objects adhere to their specified attributes in the prompt. B2B is designed as a compatible plug-and-play module for existing T2I models like Stable Diffusion and Gligen, markedly enhancing models' performance in addressing these key challenges. We assess our technique on the well-established CompBench and TIFA score benchmarks, and HRS dataset where B2B not only surpasses methods specialized in either attribute binding or layout guidance but also uniquely excels by integrating these capabilities to deliver enhanced overall performance.
- Open Access
- 6 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 141313-141327
Faster Image2Video Generation: A Closer Look at CLIP Image Embedding’s Impact on Spatio-Temporal Cross-Attentions
This paper explores the effectiveness—specifically in improving video consistency—and the computational burden of Contrastive Language-Image Pre-Training (CLIP) embeddings in video generation. The investigation is conducted using the Stable Video Diffusion (SVD) framework, a state-of-the-art method for generating high-quality videos from image inputs. The diffusion process in SVD generates videos by iteratively denoising noisy inputs over multiple steps. Our analysis reveals that employing CLIP in the cross-attention mechanism at every step of this denoising process has limited impact on maintaining subject and background consistency while imposing a significant computational burden on the video generation network. To address this, we propose Video Computation Cut (VCUT), a novel, training-free optimization method that significantly reduces computational demands without compromising output quality. VCUT replaces the computationally intensive temporal cross-attention with a one-time computed linear layer, cached and reused across inference steps. This innovation reduces up to 322T MACs per 25-frame video, decreases model parameters by 50M, and cuts latency by 20% compared to baseline methods. By streamlining the SVD architecture, our approach makes high-quality video generation more accessible, cost-effective, and eco-friendly, paving the way for real-time applications in telemedicine, remote learning, and automated content creation.
- 6 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 141313-141327
Open AccessFaster Image2Video Generation: A Closer Look at CLIP Image Embedding’s Impact on Spatio-Temporal Cross-Attentions
This paper explores the effectiveness—specifically in improving video consistency—and the computational burden of Contrastive Language-Image Pre-Training (CLIP) embeddings in video generation. The investigation is conducted using the Stable Video Diffusion (SVD) framework, a state-of-the-art method for generating high-quality videos from image inputs. The diffusion process in SVD generates videos by iteratively denoising noisy inputs over multiple steps. Our analysis reveals that employing CLIP in the cross-attention mechanism at every step of this denoising process has limited impact on maintaining subject and background consistency while imposing a significant computational burden on the video generation network. To address this, we propose Video Computation Cut (VCUT), a novel, training-free optimization method that significantly reduces computational demands without compromising output quality. VCUT replaces the computationally intensive temporal cross-attention with a one-time computed linear layer, cached and reused across inference steps. This innovation reduces up to 322T MACs per 25-frame video, decreases model parameters by 50M, and cuts latency by 20% compared to baseline methods. By streamlining the SVD architecture, our approach makes high-quality video generation more accessible, cost-effective, and eco-friendly, paving the way for real-time applications in telemedicine, remote learning, and automated content creation.
- Open Access
- 19 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 44 (6) , 2657-2671
Organ-DETR: Organ Detection via Transformers
Query-based Transformers have been yielding impressive performance in object localization and detection tasks. However, their application to organ detection in 3D medical imaging data has been relatively unexplored. This study introduces Organ-DETR, featuring two innovative modules, MultiScale Attention (MSA) and Dense Query Matching (DQM), designed to enhance the performance of Detection Transformers (DETRs) for 3D organ detection. MSA is a novel top-down representation learning approach for efficiently encoding Computed Tomography (CT) features. This architecture employs a multiscale attention mechanism, utilizing both dual self-attention and cross-scale attention mechanisms to extract intra- and inter-scale spatial interactions in the attention mechanism. Organ-DETR also introduces DQM, an approach for one-to-many matching that tackles the label assignment difficulties in organ detection. DQM increases positive queries to enhance both recall scores and training efficiency without the need for additional learnable parameters. Extensive results on five 3D CT datasets indicate that the proposed Organ-DETR outperforms comparable techniques by achieving a remarkable improvement of +10.6 mAP COCO. The project and code are available at https://github.com/ai-med/ OrganDETR.
- 19 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 44 (6) , 2657-2671
Open AccessOrgan-DETR: Organ Detection via Transformers
Query-based Transformers have been yielding impressive performance in object localization and detection tasks. However, their application to organ detection in 3D medical imaging data has been relatively unexplored. This study introduces Organ-DETR, featuring two innovative modules, MultiScale Attention (MSA) and Dense Query Matching (DQM), designed to enhance the performance of Detection Transformers (DETRs) for 3D organ detection. MSA is a novel top-down representation learning approach for efficiently encoding Computed Tomography (CT) features. This architecture employs a multiscale attention mechanism, utilizing both dual self-attention and cross-scale attention mechanisms to extract intra- and inter-scale spatial interactions in the attention mechanism. Organ-DETR also introduces DQM, an approach for one-to-many matching that tackles the label assignment difficulties in organ detection. DQM increases positive queries to enhance both recall scores and training efficiency without the need for additional learnable parameters. Extensive results on five 3D CT datasets indicate that the proposed Organ-DETR outperforms comparable techniques by achieving a remarkable improvement of +10.6 mAP COCO. The project and code are available at https://github.com/ai-med/ OrganDETR.
- 5 October 2024
- Published by Springer Nature
- p. 86-97
Mamba? Catch The Hype Or Rethink What Really Helps for Image Registration
- 5 October 2024
- Published by Springer Nature
- p. 86-97
Mamba? Catch The Hype Or Rethink What Really Helps for Image Registration
- Open Access
Mamba? Catch The Hype Or Rethink What Really Helps for Image Registration
Our findings indicate that adopting "advanced" computational elements fails to significantly improve registration accuracy. Instead, well-established registration-specific designs offer fair improvements, enhancing results by a marginal 1.5\% over the baseline. Our findings emphasize the importance of rigorous, unbiased evaluation and contribution disentanglement of all low- and high-level registration components, rather than simply following the computer vision trends with "more advanced" computational blocks. We advocate for simpler yet effective solutions and novel evaluation metrics that go beyond conventional registration accuracy, warranting further research across diverse organs and modalities. The code is available at \url{https://github.com/BailiangJ/rethink-reg}.
Open AccessMamba? Catch The Hype Or Rethink What Really Helps for Image Registration
Our findings indicate that adopting "advanced" computational elements fails to significantly improve registration accuracy. Instead, well-established registration-specific designs offer fair improvements, enhancing results by a marginal 1.5\% over the baseline. Our findings emphasize the importance of rigorous, unbiased evaluation and contribution disentanglement of all low- and high-level registration components, rather than simply following the computer vision trends with "more advanced" computational blocks. We advocate for simpler yet effective solutions and novel evaluation metrics that go beyond conventional registration accuracy, warranting further research across diverse organs and modalities. The code is available at \url{https://github.com/BailiangJ/rethink-reg}.
- Open Access
- 10 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Imaging
- Vol. 10 (25730436) , 1094-1110
No-Clean-Reference Image Super-Resolution: Application to Electron Microscopy
The inability to acquire clean high-resolution (HR) electron microscopy (EM) images over a large brain tissue volume hampers many neuroscience studies. To address this challenge, we propose a deep-learning-based image super-resolution (SR) approach to computationally reconstruct a clean HR 3D-EM image with a large field of view (FoV) from noisy low-resolution (LR) acquisition. Our contributions are I) investigation of training with no-clean references; II) introduction of a novel network architecture, named EMSR, for enhancing the resolution of LR EM images while reducing inherent noise. The EMSR leverages distinctive features in brain EM images–repetitive textural and geometrical patterns amidst less informative backgrounds– via multiscale edge-attention and self-attention mechanisms to emphasize edge features over the background; and, III) comparison of different training strategies including using acquired LR and HR image pairs, i.e., real pairs with no-clean references contaminated with real corruptions, pairs of synthetic LR and acquired HR, as well as acquired LR and denoised HR pairs. Experiments with nine brain datasets showed that training with real pairs can produce high-quality super-resolved results, demonstrating the feasibility of training with nonclean references. Additionally, comparable results were observed, both visually and numerically, when employing denoised and noisy references for training. Moreover, utilizing the network trained with synthetically generated LR images from HR counterparts proved effective in yielding satisfactory SR results, even in certain cases, outperforming training with real pairs. The proposed SR network was compared quantitatively and qualitatively with several established SR techniques, demonstrating either the superiority or competitiveness of the proposed method in recovering fine details while mitigating noise.
- 10 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Imaging
- Vol. 10 (25730436) , 1094-1110
Open AccessNo-Clean-Reference Image Super-Resolution: Application to Electron Microscopy
The inability to acquire clean high-resolution (HR) electron microscopy (EM) images over a large brain tissue volume hampers many neuroscience studies. To address this challenge, we propose a deep-learning-based image super-resolution (SR) approach to computationally reconstruct a clean HR 3D-EM image with a large field of view (FoV) from noisy low-resolution (LR) acquisition. Our contributions are I) investigation of training with no-clean references; II) introduction of a novel network architecture, named EMSR, for enhancing the resolution of LR EM images while reducing inherent noise. The EMSR leverages distinctive features in brain EM images–repetitive textural and geometrical patterns amidst less informative backgrounds– via multiscale edge-attention and self-attention mechanisms to emphasize edge features over the background; and, III) comparison of different training strategies including using acquired LR and HR image pairs, i.e., real pairs with no-clean references contaminated with real corruptions, pairs of synthetic LR and acquired HR, as well as acquired LR and denoised HR pairs. Experiments with nine brain datasets showed that training with real pairs can produce high-quality super-resolved results, demonstrating the feasibility of training with nonclean references. Additionally, comparable results were observed, both visually and numerically, when employing denoised and noisy references for training. Moreover, utilizing the network trained with synthetically generated LR images from HR counterparts proved effective in yielding satisfactory SR results, even in certain cases, outperforming training with real pairs. The proposed SR network was compared quantitatively and qualitatively with several established SR techniques, demonstrating either the superiority or competitiveness of the proposed method in recovering fine details while mitigating noise.
- Open Access
Self-Supervised Super-Resolution Approach for Isotropic Reconstruction of 3D Electron Microscopy Images from Anisotropic Acquisition
Three-dimensional electron microscopy (3DEM) is an essential technique to investigate volumetric tissue ultra-structure. Due to technical limitations and high imaging costs, samples are often imaged anisotropically, where resolution in the axial direction ($z$) is lower than in the lateral directions $(x,y)$. This anisotropy 3DEM can hamper subsequent analysis and visualization tasks. To overcome this limitation, we propose a novel deep-learning (DL)-based self-supervised super-resolution approach that computationally reconstructs isotropic 3DEM from the anisotropic acquisition. The proposed DL-based framework is built upon the U-shape architecture incorporating vision-transformer (ViT) blocks, enabling high-capability learning of local and global multi-scale image dependencies. To train the tailored network, we employ a self-supervised approach. Specifically, we generate pairs of anisotropic and isotropic training datasets from the given anisotropic 3DEM data. By feeding the given anisotropic 3DEM dataset in the trained network through our proposed framework, the isotropic 3DEM is obtained. Importantly, this isotropic reconstruction approach relies solely on the given anisotropic 3DEM dataset and does not require pairs of co-registered anisotropic and isotropic 3DEM training datasets. To evaluate the effectiveness of the proposed method, we conducted experiments using three 3DEM datasets acquired from brain. The experimental results demonstrated that our proposed framework could successfully reconstruct isotropic 3DEM from the anisotropic acquisition.Open AccessSelf-Supervised Super-Resolution Approach for Isotropic Reconstruction of 3D Electron Microscopy Images from Anisotropic Acquisition
Three-dimensional electron microscopy (3DEM) is an essential technique to investigate volumetric tissue ultra-structure. Due to technical limitations and high imaging costs, samples are often imaged anisotropically, where resolution in the axial direction ($z$) is lower than in the lateral directions $(x,y)$. This anisotropy 3DEM can hamper subsequent analysis and visualization tasks. To overcome this limitation, we propose a novel deep-learning (DL)-based self-supervised super-resolution approach that computationally reconstructs isotropic 3DEM from the anisotropic acquisition. The proposed DL-based framework is built upon the U-shape architecture incorporating vision-transformer (ViT) blocks, enabling high-capability learning of local and global multi-scale image dependencies. To train the tailored network, we employ a self-supervised approach. Specifically, we generate pairs of anisotropic and isotropic training datasets from the given anisotropic 3DEM data. By feeding the given anisotropic 3DEM dataset in the trained network through our proposed framework, the isotropic 3DEM is obtained. Importantly, this isotropic reconstruction approach relies solely on the given anisotropic 3DEM dataset and does not require pairs of co-registered anisotropic and isotropic 3DEM training datasets. To evaluate the effectiveness of the proposed method, we conducted experiments using three 3DEM datasets acquired from brain. The experimental results demonstrated that our proposed framework could successfully reconstruct isotropic 3DEM from the anisotropic acquisition. - Open Access
Principal Component Wavelet Networks for Solving Linear Inverse Problems
In this paper we propose a novel learning-based wavelet transform and demonstrate its utility as a representation in solving a number of linear inverse problems—these are asymmetric problems, where the forward problem is easy to solve, but the inverse is difficult and often ill-posed. The wavelet decomposition is comprised of the application of an invertible 2D wavelet filter-bank comprising symmetric and anti-symmetric filters, in combination with a set of convolution filters learnt from Principal Component Analysis (PCA). The filters are needed to control the size of the decomposition. We show that the application of PCA across wavelet subbands in this way produces an architecture equivalent to a separable Convolutional Neural Network (CNN), with the principal components forming the filters and the subtraction of the mean forming the bias terms. The use of an invertible filter bank and (approximately) invertible PCA allows us to create a deep autoencoder very simply, and avoids issues of overfitting. We investigate the construction and learning of such networks, and their application to linear inverse problems via the Alternating Direction of Multipliers Method (ADMM). We use our network as a drop-in replacement for traditional discrete wavelet transform, using wavelet shrinkage as the projection operator. The results show good potential on a number of inverse problems such as compressive sensing, in-painting, denoising and super-resolution, and significantly close the performance gap with Generative Adversarial Network (GAN)-based methods.Open AccessPrincipal Component Wavelet Networks for Solving Linear Inverse Problems
In this paper we propose a novel learning-based wavelet transform and demonstrate its utility as a representation in solving a number of linear inverse problems—these are asymmetric problems, where the forward problem is easy to solve, but the inverse is difficult and often ill-posed. The wavelet decomposition is comprised of the application of an invertible 2D wavelet filter-bank comprising symmetric and anti-symmetric filters, in combination with a set of convolution filters learnt from Principal Component Analysis (PCA). The filters are needed to control the size of the decomposition. We show that the application of PCA across wavelet subbands in this way produces an architecture equivalent to a separable Convolutional Neural Network (CNN), with the principal components forming the filters and the subtraction of the mean forming the bias terms. The use of an invertible filter bank and (approximately) invertible PCA allows us to create a deep autoencoder very simply, and avoids issues of overfitting. We investigate the construction and learning of such networks, and their application to linear inverse problems via the Alternating Direction of Multipliers Method (ADMM). We use our network as a drop-in replacement for traditional discrete wavelet transform, using wavelet shrinkage as the projection operator. The results show good potential on a number of inverse problems such as compressive sensing, in-painting, denoising and super-resolution, and significantly close the performance gap with Generative Adversarial Network (GAN)-based methods. - Open Access
- 24 March 2021
- Published by Frontiers Media SA in Frontiers in Plant Science
- Vol. 12
Deep Segmentation of Point Clouds of Wheat
The 3D analysis of plants has become increasingly effective in modeling the relative structure of organs and other traits of interest. In this paper, we introduce a novel pattern-based deep neural network, Pattern-Net, for segmentation of point clouds of wheat. This study is the first to segment the point clouds of wheat into defined organs and to analyse their traits directly in 3D space. Point clouds have no regular grid and thus their segmentation is challenging. Pattern-Net creates a dynamic link among neighbors to seek stable patterns from a 3D point set across several levels of abstraction using the K-nearest neighbor algorithm. To this end, different layers are connected to each other to create complex patterns from the simple ones, strengthen dynamic link propagation, alleviate the vanishing-gradient problem, encourage link reuse and substantially reduce the number of parameters. The proposed deep network is capable of analysing and decomposing unstructured complex point clouds into semantically meaningful parts. Experiments on a wheat dataset verify the effectiveness of our approach for segmentation of wheat in 3D space.
- 24 March 2021
- Published by Frontiers Media SA in Frontiers in Plant Science
- Vol. 12
Open AccessDeep Segmentation of Point Clouds of Wheat
The 3D analysis of plants has become increasingly effective in modeling the relative structure of organs and other traits of interest. In this paper, we introduce a novel pattern-based deep neural network, Pattern-Net, for segmentation of point clouds of wheat. This study is the first to segment the point clouds of wheat into defined organs and to analyse their traits directly in 3D space. Point clouds have no regular grid and thus their segmentation is challenging. Pattern-Net creates a dynamic link among neighbors to seek stable patterns from a 3D point set across several levels of abstraction using the K-nearest neighbor algorithm. To this end, different layers are connected to each other to create complex patterns from the simple ones, strengthen dynamic link propagation, alleviate the vanishing-gradient problem, encourage link reuse and substantially reduce the number of parameters. The proposed deep network is capable of analysing and decomposing unstructured complex point clouds into semantically meaningful parts. Experiments on a wheat dataset verify the effectiveness of our approach for segmentation of wheat in 3D space.