Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Georgios TziritasSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works190 from public data
- Sequential Deep Learning for Human Action Recognition817
A fully automated deep model, which learns to classify human actions without using any prior knowledge is proposed, which outperforms existing deep models, and gives comparable results with the best related works.
- The 2005 PASCAL Visual Object Classes Challenge742
This chapter provides details of the datasets, algorithms used by the teams, evaluation criteria, and results achieved in the PASCAL Visual Object Classes Challenge.
- Face detection using quantized skin color regions merging and wavelet packet analysis670
An efficient and reliable probabilistic metric derived from the Bhattacharrya distance is used in order to classify the extracted feature vectors into face or nonface areas, using some prototype face area vectors, acquired in a previous training stage.
- 607
- The Visual Object Tracking VOT2014 Challenge Results588
The Visual Object Tracking challenge 2014, VOT2014, aims at comparing short-term single-object visual trackers that do not apply pre-learned models of object appearance and introduces a new dataset with full annotation of targets by rotated bounding boxes and per-frame attribute.
- Machine Learning Challenges. Evaluating Predictive Uncertainty, Visual Object Classification, and Recognising Tectual Entailment210
A Bayesian approach to Predictive Uncertainty and a Lexical Alignment Model for Probabilistic Textual Entailment are presented.
- Wavelet packet analysis for face recognition161
It is shown how an efficient and reliable probabilistic metric derived from the Bhattacharrya distance can be used in order to classify the face feature vectors into person classes.
- Simplifying ConvNets for Fast Learning144
This paper proposes different strategies for simplifying filters, used as feature extractors, to be learnt in convolutional neural networks (ConvNets) in order to modify the hypothesis space, and to speed-up learning and processing times.
- 136
- Action Classification in Soccer Videos with Long Short-Term Memory Recurrent Neural Networks130
Experimental results show that the proposed approach for action classification in soccer videos outperforms classification methods of related works, and that the combination of the two features (BoW and dominant motion) leads to a classification rate of 92%.
- 117
- 100
- 98
- Spatio-Temporal Convolutional Sparse Auto-Encoder for Sequence Classification84
A novel learning-based approach for video sequence classification that automatically learns a sparse shift-invariant representation of the local 2D+t salient information, without any use of prior knowledge is presented.
- Evaluation of video activity localizations integrating quality and quantity measurements83
A new performance metric addressing and unifying the qualitative and quantitative aspects of the performance measures is proposed, which has been tested on several activity recognition algorithms participating in the ICPR 2012 HARL competition.
- BLSTM-RNN Based 3D Gesture Classification80
A new robust method for inertial MEM (MicroElectroMechanical systems) 3D gesture recognition based on Bidirectional Long Short-Term Memory Recurrent Neural Networks (BLSTM-RNN) for gesture classification from raw MEM data is presented.
- TEXT DETECTION WITH CONVOLUTIONAL NEURAL NETWORKS80
This work presents an approach based on convolutional neural networks to detect and localize horizontal text lines from raw color pixels and demonstrates that it can outperform other methods on the real-world test set of ICDAR’03.
- Deep Model Compression and Architecture Optimization for Embedded Systems: A Survey79
A survey of methods suitable for porting deep neural networks on resource-limited devices, especially for smart cameras, and introduces the methods to enhance networks structures as well as neural architecture search techniques.
- A Wavelet-based Framework for Face Recognition75
A new scheme for face recognition using a wavelet packet decomposition is presented, and an efficient and reliable probalistic metric derived from the Bhattacharrya distance is used in order to classify the face feature vectors into person classes.
- Bayesian Level Sets for Image Segmentation73
A level set formulation is used to model the boundaries of the image regions and a new Multilabel Fast Marching is introduced for the evolution of the region contours toward the segmentation result, leading to the Bayesian Level Set methodology.
- 70
- Pose Estimation using Point and Line Correspondences68
This paper presents two fast methods for estimating the 3D pose using 2D to 3D point and line correspondences and concludes that the iterative paraperspective method has better convergence properties than the iteratively improved weak perspective method.
- Siamese multi-layer perceptrons for dimensionality reduction and face identification59
While the classical MLP fixes the dimension of the output space, the siamese MLP allows flexible output dimension, hence it is applied for visualization of the dimensionality reduction to the 2-d and 3-d spaces.
- Contribution of recurrent connectionist language models in improving LSTM-based Arabic text recognition in videos58
This work introduces recurrent connectionist language modeling in order to improve Long-Short Term Memory (LSTM) based Arabic text recognition in videos, and proposes a modified beam search algorithm that uses both OCR and language model probabilities in parallel at each decoding time-step.
- 57
- Low-Complexity Approximate Convolutional Neural Networks51
The idea is to approximate all elements of a given ConvNet and replace the original convolutional filters and parameters with an efficient approximations capable of extreme reductions in computational complexity.
- 46
- An Online Backpropagation Algorithm with Validation Error-Based Adaptive Learning Rate46
The proposed algorithm is a heuristic method consisting of two phases that rapidly converges and that it outperforms standard Backpropagation in terms of generalization when the size of the training set is reduced.
- 44
- 43
- Class-balanced siamese neural networks40
This paper focuses on metric learning with Siamese Neural Networks (SNN) and proposes three contributions: a tuple-based architecture, an objective function with a norm regularisation and a polar sine-based angular reformulation for cosine dissimilarity learning.
- 40
- 39
- 38
- Industry and Object Recognition: Applications, Applied Research and Challenges38
Object recognition technology has matured to a point at which exciting applications are becoming possible, and prospects for future applications and developments in industry are discussed.
- 37
- Local-feature-map integration using convolutional neural networks for music genre classification37
This paper focuses on appropriate feature extraction and proper classification by integrating the features using convolutional neural network, and calculates gray level co-occurrence matrix (GLCM) descriptors with different offsets from a short-term mel spectrogram.
- 36
- Inertial Gesture Recognition with BLSTM-RNN36
Experimental results show that the proposed approach outperforms classical methods with an average classification rate of 95.57% and a standard deviation of 0.50 for 616 test gestures, and underline that combining accelerometer and gyrometer data gives better results than using a single inertial description.
- Supervised Learning and Codebook Optimization for Bag-of-Words Models35
A new supervised method for joint codebook creation and class learning, which learns the cluster centers of the codebook in a goal-directed way using the class labels of the training set, which allows the discriminative power of an unsupervised learned codebook to be kept or to be improved, thus decreasing the computational complexity due to the nearest neighbor search.
- 34
- 30
- A Neural Scheme for Robust Detection of Transparent Logos in TV Programs30
This paper presents a connectionist approach for detecting and precisely localizing transparent logos in TV programs, that does not require any local preprocessing and leads to a straightforward real time implementation.
- 29
- Resolution enhancement of textual images via multiple coupled dictionaries and adaptive sparse representation selection28
Significant improvements in visual quality and character recognition rates are achieved using the proposed approach, confirmed by a detailed comparative study with state-of-the-art upscaling approaches.
- Robust camera calibration using 2D-to-3D feature correspondences28
A method for optimally estimating the internal and external camera parameters from point or line correspondences, and an elegant way to automatically establish 2D to 3D feature correspondences using projective geometry are introduced.
- 27
- Resolution enhancement of textual images: a survey of single image‐based methods26
This study surveys methods that are mainly designed for enhancing low-resolution textual images in super-resolution (SR) task and criticises these methods and discusses areas which promise improvements in such task.
- A comprehensive neural-based approach for text recognition in videos using natural language processing26
A complete video Optical Character Recognition system, specifically adapted to detect and recognize embedded texts in videos, that achieves an outstanding character recognition rate of 95%, corresponding to 78% of words correctly recognized, which enables its incorporation into an automatic video indexing and retrieval system.
- Improving texture categorization with biologically-inspired filtering25
A novel, simple, yet very powerful biologically-inspired filtering (BF) which simulates the performance of human retina is proposed which improves significantly theperformance of many texture classification systems, notably in noisy environments.
- Text recognition in multimedia documents: a study of two neural-based OCRs using and avoiding character segmentation23
Two neural-based optical character recognition systems that handle the text recognition problem in different ways are proposed, with outstanding results showing that the proposed approaches outperform the state-of-the-art methods.
- 22
- 22
- A Comparison between Multi-Layer Perceptrons and Convolutional Neural Networks for Text Image Super-Resolution21
An example-based framework for both MLP and ConvNet, where a non-linear mapping between pairs of patches and high-frequency pixel values is learned, which demonstrates that for equivalent complexity, ConvNets are better than MLPs at predicting missing details in upsampled text images.
- Real-Time Video Convolutional Face Finder on Embedded Platforms21
Experimental results show that the embedded face detection system can accurately locate faces with less computational load and memory cost, and runs on a 275 MHz Starcore DSP at 35 QCIF images/s with state-of-the-art detection rates and very low false alarm rates.
- TWO-DIMENSIONAL-ORIENTED LINEAR DISCRIMINANT ANALYSIS FOR FACE RECOGNITION21
Two-Dimensional- Oriented Linear Discriminant Analysis is presented and experimental results indicate that 2DO-LDA is more efficient than both 2D-PCA and LDA when dealing with variations in lighting conditions, facial expression and head pose.
- Comparing Robustness of Two-Dimensional PCA and Eigenfaces for Face Recognition21
This paper determines, for each preprocessing step, the minimum accuracy required in order to allow successful face recognition with 2D-PCA and compares it with the classical eigenfaces method.
- Comparing Robustness of Two-Dimensional PCA and Eigenfaces for Face Recognition Springer21
This paper determines, for each preprocessing step, the minimum accuracy required in order to allow successful face recognition with 2D-PCA and compares it with the classical eigenfaces method.
- 20
- 2D Wasserstein loss for robust facial landmark detection19
A novel method for robust facial landmark detection using a loss function based on the 2D Wasserstein distance combined with a new landmark coordinate sampling relying on the barycenter of the individual propability distributions is proposed.
- Scene Segmentation and Image Feature Extraction for Video Indexing and Retrieval19
A video analysis and indexing engine, that can perform fully automatic scene segmentation and feature extraction, in the context of a television archive, based on a library of image analysis functions and templates is presented.
- 18
- 18
- Enhanced sports broadcasting by means of augmented reality in MPEG-418
The architecture of an application that will address the needs of viewers and broadcasters for enhanced content creation, viewing and interaction by means of augmented reality for sports broadcasting is presented, carried out in the framework of the IST project PISTE funded by the European Commission.
- An automatic inspection system for the detection of tire surface defects and their severity classification through a two-stage multimodal deep learning approach17
An innovative multimodal approach aimed at automating the tire quality control process through the use of deep learning on data obtained from stereo-photometric cameras meticulously integrated into a purpose-built, sophisticated tire acquisition system capable of comprehensive data capture across all tire zones.
- 16
- A high‐performance face detection system using OpenMP16
The presented system is able to process images in real time (38 images/sec) by sustaining almost linear speedups on a system with a quad‐core processor and a particular OpenMP runtime library.
- 16
- 15
- 15
- 15
- ROBUST FACE ALIGNMENT USING CONVOLUTIONAL NEURAL NETWORKS14
A novel technique based on a specific neural architecture which precisely aligns face images extracted from bounding boxes coming from a face detector without localizing any facial feature points is presented.
- 14
- DISTANCE MAPS: A ROBUST ILLUMINATION PREPROCESSING FOR ACTIVE APPEARANCE MODELS14
A robust preprocessing method designed to enhance the performances of deformable models methods in the case of lighting variations is applied to the Active Appearance Models (AAM), which consists in replacing texture images by distance maps as input of the deformable appearance models methods.
- Online face detection and user authentication14
This technical demo shows the latest image processing techniques for face detection developed at France Telecom and for face recognition developed at Eurécom.
- Virtual meeting in cyberstage14
The scientific approach involves stereo-camera calibration and rectification, and use of the camera parameters for integrating the stereovideo into a virtual environment, while maintaining the stereo-effect and correct perspective for each participant.
- RESIST: Robust Transformer for Unsupervised Time Series Anomaly Detection13
This paper introduces a robust learning strategy that trains a Transformer to model the nominal behaviour of the network activity, and proposes RESIST, a Robust transformEr developed for unSupervised tIme Series anomaly deTection.
- Convolutive Bottleneck Network with Dropout for Dysarthric Speech Recognition13
A robust feature extraction method using a convolutive bottleneck network (CBN) instead of the well-known MFCC to recognition of speech produced by a person with an articulation disorder resulting from athetoid cerebral palsy is proposed.
- 13
- 13
- Fine-grained facial landmark detection exploiting intermediate feature representations12
A new approach for Fine-grained Facial Landmark Detection (FFLD) improving on the precision of the detected points is presented, and extensive experiments show that the framework significantly increases the local precision of state-of-the-art deep coordinate regression models.
- Handling noise in textual image resolution enhancement using online and offline learned dictionaries12
This study proves that the application of an efficient sparse coding-based denoising process followed by the magnification process can achieve good restoration results even if the input image is highly noisy.
- 12
- 12
- Sensor fusion for social robotics12
The paper presented at the 31st International Symposium on Robotics (ISR2000) at the Palais des congres, Montreal, Canada, presents a probabilistic method for calculating the E-modulus of the response of the immune system to mechanical forces.
- Learning Sparse Filters in Deep Convolutional Neural Networks with a $$l_1/l_2$$ Pseudo-Norm11
This paper presents a sparsity-inducing regularization term based on the ratio l1/l2 pseudo-norm defined on the filter coefficients, which significantly reduces the number of filters of classical models such as LeNet and VGG while reaching the same or even better accuracy than the baseline models.
- 11
- 11
- 11
- 11
- Learning personalized ADL recognition models from few raw data10
This paper proposes an approach to learn personalized ADL recognition models from few raw data based on a specific type of neural network called matching network, which can generalize easily to new activities without needing extra training, thus making them very versatile for real applications.
- Two-stage human hair segmentation in the wild using deep shape prior10
A novel hair segmentation approach integrating a deep shape prior into a carefully designed two-stage Fully Convolutional Neural Network pipeline that achieves state-of-the-art performance on the LFW-Part and Figaro1k datasets.
- A survey of deep facial landmark detection10
A survey of deep facial landmark detection for 2D images and video and a comparative analysis of different face alignment approaches is provided as well as some future research directions.
- 10
- Single Textual Image Super-Resolution Using Multiple Learned Dictionaries Based Sparse Coding10
The proposed approach is able to build more representative dictionaries learned from a large training Low-Resolution/High-Res resolution patch pair database and outperforms its counterparts in visual fidelity as well as in numerical measures.
- Text Recognition in Videos Using a Recurrent Connectionist Approach10
This paper proposes a novel approach able to avoid any explicit character segmentation in OCR systems developed to recognize texts embedded in multimedia documents using a multi-scale scanning scheme that achieves very high recognition rates.
- Adapted Active Appearance Models10
This work proposes Adapted Active Appearance Models to align unknown faces in unknown poses and illuminations based on a specific transformation of the active model texture in an oriented map, which changes the AAM normalization process.
- Modeling gaze behavior for a 3D ECA in a dialogue situation10
The ECA's gaze control results from the fusion of a rational dialogue engine based on natural language interaction and a multi-users face tracker.
- WaveRead: Automatic measurement of relative gene expression levels from microarrays using wavelet analysis10
WaveRead is described, a fully automated, standalone, open-source system for quantifying gene expression array images that is able to automatically grid the image and quantify signal intensities and background corrections without any user intervention.
- Improving Information Extraction on Business Documents with Specific Pre-training Tasks9
LayoutLM, a language model pre-trained on a collection of business documents, is used, and a new post-processing algorithm to decode BIESO tags in Information Extraction that performs better with complex entities is introduced.
- 9
- 9
- An Automatic Method for Video Character Segmentation9
An automatic segmentation system for characters in text color images cropped from natural images or videos based on a new neuronal architecture insuring fast processing and robustness against noise, variations in illumination, complex background and low resolution is presented.
- 8
- Exploiting Contextual Motion Cues for Visual Object Tracking8
An algorithm for on-line, real-time tracking of arbitrary objects in videos from unconstrained environments is proposed based on a particle filter framework using different visual features and motion prediction models and proposes a new method to collect negative training examples for updating the classifier at each video frame.
- 8
- Bilinear Discriminant Analysis for Face Recognition8
The experimental results indicate that BDA outperforms RoLDA, CoLDA and 2DPCA for face recognition, while leading to a significant dimensionality reduction.
- Fully Vision-based Calibration of a Hand-Eye Robot8
This article is concerned with calibrating an anthropomorphic two-armed robot equipped with a stereo-camera vision system, that is estimating the different geometric relationships involved in the model of the robot.
- 7
- 7
- Similarity Metric Learning7
An overview of the principle methods and models used for similarity metric learning with neural networks is given, describing the most common architectures, loss functions and training algorithms.
- On the impact of outliers on high-dimensional data analysis methods for face recognition7
A filtering process is proposed that allows an automatic isolation of noisy faces which are responsible for the performance degradation and is based-on the recently proposed robust high-dimensional data analysis method RobPCA.
- 6
- Robust detection of outliers for projection-based face recognition methods6
A filtering process is proposed that allows the automatic selection of noisy face images which are responsible for the performance degradation and significantly improves recognition rates by 10 to 30%.
- Optimal projection of 2-D displacements for 3-D translational motion estimation6
A novel method based on an unbiased estimator that aims at enhancing stability in 3-D motion parameters from 2-D displacements and strongly reduces the influence of noise contamination is presented.
- Robust Face Detection Based on Convolutional Neural Networks6
A connectionist approach for detecting and precisely localizing semi-frontal human faces in complex images, making no assumption on the content or the lighting conditions of the scene, neither on the size, the orientation, and the appearance of the faces.
- Routine Modeling with Time Series Metric Learning5
Experimental results show that a clustering algorithm provided with the learned distance is able to recover daily routines, and an architecture based on sequence-to-sequence models is proposed to learn a distance between time series.
- 5
- 5
- Face Recognition using Modular Bilinear Discriminant Analysis5
Modular Bilinear Discriminant Analysis is presented, a new approach for face recognition that is more effective than both BDA and the Modular Eigenspaces approach forFace recognition.
- Meet.Me@Cyberstage: towards Immersive Telepresence5
Today’s technology and advances in telecommunication rapidly change the way the business is carried out, making it a globally distributed process, in which communication and collaboration of geographically dispersed groups is of vital importance, and VR systems are adapting accordingly, by providing not only a better man-machine interface, but also by facilitating human to human interaction and collaboration over distance.
- Long-Range Transformer Architectures for Document Understanding4
This paper introduces 2 new multi-modal (text + layout) long-range models for DU based on efficient implementations of Transformers for long sequences, and proposes 2D relative attention bias to guide self-attention towards relevant tokens without harming model efficiency.
- 4
- AColDPS - Robust and Unsupervised Automatic Color Document Processing System4
This paper presents the first fully automatic color analysis system suited for business documents that uses mainly color morphology and outperforms the classical approach that uses binarization of greyscale images.
- 4
- 4
- A Combination of Hand-Crafted and Hierarchical High-Level Learnt Feature Extraction for Music Genre Classification4
A new approach for automatic music genre classification which relies on learning a feature hierarchy with a deep learning architecture over hand-crafted feature extracted from an audio signal that clearly outperforms the state-of-the-art approaches.
- 4
- Conceptual Indexing of Television Images Based on Face and Caption Sizes and Locations4
This paper presents some methods for image classification in video, based on the presence, size and location of faces and captions, and argues that such classifications are highly domain-dependent and are best handled using flexible knowledge management systems.
- 3
- 3
- 3
- 3
- Blind Super-Resolution with Deep Convolutional Neural Networks3
This paper addresses blind SR using Convolutional Neural Networks and shows that such models can handle different level of blur without any a priori knowledge of the actual kernel used to produce LR images.
- Histogram of Structure Tensors: Application to Pattern Clustering3
The proposed HoST descriptor succeeds in improving the performance of clustering based resolution enhancement approaches and the use of the histogram concept makes the proposed descriptor so effective and useful when a reduced feature representation is required.
- EFFICIENT OBJECT DETECTION ROBUST TO RST WITH MINIMAL SET OF EXAMPLES3
This paper presents a detection system that first searches the likely positions and scales of the object using image preprocessing and cross-correlation method and secondly, uses a similarity measure based on affine deformation to confirm or not the predetection.
- 3
- 2
- 2
- Unsupervised learning of co-occurrences for face images retrieval2
This work proposes a framework that substantially improves face retrieval by exploiting the faces occurring simultaneously in a query's context to infer a multi-dimensional social context descriptor and combining this compact structural descriptor with the individual visual face features in a common feature vector considerably increases the correct face retrieval rate.
- 2
- Polar Sine Based Siamese Neural Network for Gesture Recognition2
This work focuses on metric learning between gesture sample signatures using Siamese Neural Networks (SNN), which aims at modeling semantic relations between classes to extract discriminative features through metric learning.
- 2
- Incremental Principal Component Analysis-Based Sparse Representation for Face Pose Classification2
Experimental results show that the proposed ASRC algorithm to deal with face pose estimation in occlusion, bad illumination and low-resolution cases is very robust when the illumination condition changes very dynamically and image resolutions are quite poor.
- 2
- 2
- Avatar Puppetry Using Real-Time Audio and Video Analysis2
A system which consists of a lifelike agent animated in real-time using video and audio analysis from the user using Active Appearance Models and Hidden Markov Model for Instant Messaging.
- Object pose by affine iterations2
This method is based on the iterative use of a paraperspective camera model which is a first order approximation of perspective and is inspired by DeMenthon's method (1995) which consists of determining the pose from point correspondences.
- 1
- 1
- 1
- 1
- AColDSS: Robust Unsupervised Automatic Color Segmentation System for Noisy Heterogeneous Document Images1
The first fully automatic color analysis system suited for noisy heterogeneous documents and the first fully data-driven pixel-based approach that does not need a priori information, training or manual assistance is presented.
- Boosting-based approaches for Arabic text detection in news videos1
Two boosting-based approaches for Arabic embedded text detection in news videos using Multi-Block Local Binary Patterns features and a multiexit asymmetric boosting cascade are proposed.
- Exploring Interest Points and Local Descriptors for Word Spotting Application on Historical Handwriting Images1
This paper proposes an investigation on the description of handwriting by applying different interest points and local descriptors on historical handwritten images in the context of a coarse-to-fine segmentation-free word spotting method and the observation and analysis based on the experimental results can help optimizing thedescription of handwriting according to different applications.
- Kernel Similarity Based AAMs for Face Recognition1
A Kernel similarity based Active Appearance Models (KSAAMs) is presented in which a Kernel Method is used to replace Principal Component Analysis (PCA) which is used for feature extraction in Active Appearance models.
- 1
- Sélection de singularités locales par stimulation de carte GHSOM1
This paper designs a scheme that relies on a Growing Hierarchical Self-Organizing Map (GHSOM) and obtains persuasive results with more than 82% of classification rate and offers more than 95% of good identifications.
- 1
- RAViT: Resolution-Adaptive Vision Transformer–
A novel framework for image classification called RAViT based on a multi-branch network that operates on several copies of the same image with different resolutions to reduce the computational cost while preserving the overall accuracy.
- –
- –
- –
- –
- –
- –
- –
- Online Appearance-Motion Coupling for Multi-Person Tracking in Videos–
This paper proposes a solution by online detecting model failures and accordingly adjusting the coupling between appearance and motion models, letting the functional models take over when certain model faces data association ambiguity, and at the same time suppressing the influence of inappropriate observations during model update.
- –
- –
- –
- –
- Représentation des mots manuscrits par graphe pour la recherche par similarité–
Dans ce papier, nous proposons une nouvelle approche de la recherche de mots par similarite reposant sur une structure de graphes integrant des informations sur the topologie, the morphologie locale des mots ainsi que des informationations contextuelles du voisinage de chaque point d’interet.
- –
- Probabilistic Active Appearance Models–
A probability method based on the analysis of the correlation between observed variables, which can better follow the distribution of data, to improve Principal Com- ponent Analysis which is used for feature extraction in Active Appearance Model.
- Supervised learning and codebook optimization with neural network–
A new supervised framework for joint codebook creation and class learning is proposed, which learns the codewords in a goal-directed way using the class labels of the training set, which makes the codebook becomes more discriminative.
- –
- –
- Modèles actifs d'apparences adaptés–
A robust Active Appearance Models allowing a real-time implementation and the proposed Oriented Map AAM (OM-AAM), which will be presented after to increase the AAM robustness to any other types of variability (in identity, pose, expression etc.)...
- –
- –
- –
- Efficient Codebooks for Visual Concept Recognition by ASSOms Activation–
A novel method for robustly characterizing and classifying visual concepts, and more precisely for detecting several categories of complex objects within images, that relies on Adaptive-Subspace SelfOrganizing Maps (ASSOMs).
- –
- –
- –
- –
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.