Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks294 from public data
- 736
- 681
- 669
- Deep Learning for Emotion Recognition on Small Datasets using Transfer Learning656
This paper presents the techniques employed in the team's submissions to the 2015 Emotion Recognition in the Wild contest, for the sub-challenge of Static Facial Expression Recognition In the Wild.
- Digital Video Quality: Vision Models and Metrics578
The author explains the motivation behind the work and some of the techniques used in the writing of this book, which aim to improve the quality of measurement in the field of color space perception.
- Perceptual blur and ringing metrics: application to JPEG2000575
A full- and no-reference blur metric as well as a full-reference ringing metric are presented, based on an analysis of the edges and adjacent regions in an image and have very low computational complexity.
- 544
- 465
- 331
- Perceptual distortion metric for digital color video315
A distortion metric for color video sequences is presented, based on a contrast gain control model of the human visual system that incorporates spatial and temporal aspects of vision as well as color perception.
- Synovial cyst of spinal facet292
The authors report a case in which a true synovial cyst was found attached to a cervical facet joint, and the anatomically oriented term “juxta-facet cyst” is proposed to include both ganglion and synovIAL cysts seen in this area.
- Issues in vision modeling for perceptual video quality assessment260
This paper discusses issues in vision modeling for perceptual video quality assessment (PVQA), to explain how important characteristics of the human visual system may be incorporated in vision models for PVQA, to give a brief overview of the state-of-the-art and current efforts in this field, and to outline directions for future research.
- A vision-based masking model for spread-spectrum image watermarking201
The results show that this approach facilitates the insertion of a more robust watermark while preserving the visual quality of the original, and it is demonstrated that the maximum watermark density generally does not provide the best detection performance.
- Video Quality Experts Group: current results and future directions201
The first task undertaken by VQEG was to provide a validation of objective video quality measurement methods leading to recommendations in both the telecommunications and radiocommunication sectors of the International Telecommunications Union.
- Color-Based Segmentation of Sky/Cloud Images From Ground-Based Cameras168
This paper presents a supervised segmentation framework for ground-based sky/cloud images based on a systematic analysis of different color spaces and components, using partial least-squares regression, and releases the Singapore whole Sky imaging segmentation database.
- Vision models and quality metrics for image processing applications148
These Ecole polytechnique federale de Lausanne EPFL students studied at this year's EPFL summer session studied the history of electricity and its applications in the 21st Century.
- Vision-model-based impairment metric to evaluate blocking artifacts in digital video136
Investigations are conducted to simplify and refine a vision-model-based video quality metric without compromising its prediction accuracy and the results show a strong correlation between the objective blocking ratings and the mean opinion scores on blocking artifacts.
- CloudSegNet: A Deep Network for Nychthemeron Cloud Image Segmentation127
This letter proposes a lightweight deep-learning architecture called CloudSegNet, which is the first that integrates daytime and nighttime image segmentation in a single framework and achieves state-of-the-art results on public databases.
- Color image quality on the Internet122
This paper discusses color image quality metrics and presents no-reference artifact metrics for blockiness, blurriness, and colorfulness, showing that these metrics are highly correlated with experimental data collected through subjective experiments.
- 101
- 89
- Video quality evaluation for Internet streaming applications87
In an attempt to review subjective testing procedures for video streaming applications, both Single Stimulus Continuous Quality Evaluation and Double Stimulus Impairment Scale methods are used on the same test material and an analysis of the experimental results is presented.
- Visual fidelity and perceived quality: toward comprehensive metrics83
Sharpness and colorfulness are identified among these attributes and are quantified by means of an isotropic local contrast measure and the distribution of chroma, respectively and the benefits of using these attributes are demonstrated with the help of data from subjective experiments.
- JPEG vs. JPEG 2000: an objective comparison of image encoding quality82
It is shown that the MOS predictions by the proposed tool are a better indicator of perceived image quality than PSNR, especially for highly compressed images.
- Video quality evaluation for mobile streaming applications81
The subjective ratings are used to validate the prediction performance of a real-time non-reference quality metric and compare codec performance as well as the effects of transmission errors on visual quality.
- 80
- Generation of diurnal variation for influent data for dynamic simulation77
The aim is to generate realistic influent data in terms of flow, concentrations and TKN/COD ratios and not to predict the influent of the AS plant in detail.
- 76
- Human Vision Models for Perceptually Optimized Image Processing -- A Review76
An introduction to the general issue of HVS-modeling is given and the specific applications of visual quality assessment and H VS-based image compression, which are closely related, are reviewed.
- 74
- 74
- 73
- Ground-based image analysis: A tutorial on machine-learning techniques and applications71
The advantages of using machine-learning techniques in ground-based image analysis via three primary applications: segmentation, classification, and denoising are demonstrated.
- 71
- Activated sludge modelling: development and potential use of a practical applications database69
This study analyzes a database of modelling projects to determine which biokinetic model parameters are usually changed by modellers, in which ranges, and what values are typically used for seven selected activated sludge models to help model users in the calibration step.
- Continuous monitoring in sewer networks an approach for quantification of pollution loads from CSOs into surface water bodies69
A monitoring station for continuous monitoring of load discharges from a CSO has been installed and operated for more than one year and the design and equipment of the measurement station, operational experiences and results are given.
- 68
- 66
- 64
- Motion saliency outweighs other low-level features while watching videos62
A framework for motion saliency is proposed that integrates motion vector information with spatial and temporal coherency to generate a motion attention map and finds motion to have greater influence on saliency than other low-level features when watching videos.
- 62
- 61
- Audiovisual quality evaluation of low-bitrate video60
The goals in this study are to understand the interactions between audio and video in terms of perceived audiovisual quality, and to use the subjective data to evaluate the prediction performance of the non-reference video and audio quality metrics.
- Estimating solar irradiance using sky imagers59
This paper derives and validate the model using pyranometers co-located with the authors' whole sky imagers and achieves a better performance in estimating solar irradiance and in particular its short-term variations as compared to other related methods using ground-based observations.
- Systematic study of color spaces and components for the segmentation of sky/cloud images59
This paper presents a systematic approach for the selection of color spaces and components for optimal segmentation of sky/cloud images using mainly principal component analysis (PCA) and fuzzy clustering for evaluation.
- Quality metric design: a closer look55
The first results are presented here of a number of implementation choices for some components found in most of today's visual quality metrics that are based on a model of human vision.
- PersEmoN: A Deep Network for Joint Analysis of Apparent Personality, Emotion and Their Relationship54
This paper introduces PersEmoN, an end-to-end trainable and deep Siamese-like network that consists of two convolutional network branches, one for emotion and the other for apparent personality, and explores the emotion- to-apparent-personality relationship.
- 52
- Vision-based 3D finger interactions for mixed reality games with physics simulation51
The interface proposed in this paper consists of a stereo camera, which tracks the user's hands and fingers robustly and accurately in the 3D space, and a physics engine is adopted for the simulating the physics of virtual object manipulation.
- Systematic Study of Weather Variables for Rainfall Detection50
Different weather and time-related variables - relative humidity, solar radiation, temperature, dew point, day-of-year, and time of day are analyzed systematically using Principal Component Analysis (PCA).
- Vision and Video: Models and Applications50
While traditional analog systems still form the vast majority of television sets today, production studios, broadcasters and network providers have been installing digital video equipment at an ever-increasing rate.
- Group happiness assessment using geometric features and dataset balancing48
This paper presents the techniques employed in the team's submissions to the 2016 Emotion Recognition in the Wild contest, for the sub-challenge of group-level emotion recognition, and a predominately bottom-up approach to estimate the happiness intensity of groups of people in consumer photos.
- 47
- Process simulation for recombinant protein production: Cost estimation and sensitivity analysis for heparinase I expressed inEscherichia coli47
A comparative analysis of the production costs of soluble and insoluble heparinase I expression is presented, as well as a generalized approach to sensitivity analysis, based on perturbation around a base case design scenario, to assist focusing on process steps for which improvements both are feasible and result in significant cost saving.
- Explanation of Metrizamide Brain Penetration47
Recent evidence has shown that these spaces are in fact one compartment with no diffusion barrier between them, and penetration of metrizamide into the brain is an expected rather than surprising phenomenon.
- Nighttime sky/cloud image segmentation46
The experimental results show the efficacy of the proposed algorithm for nighttime images segmentation, and the first nighttime sky/cloud image segmentation database to be released to the research community.
- Application of ion-sensitive sensors in water quality monitoring46
The function principle of ISE-sensors, their advantages, limitations and the different methods for sensor calibration are described and an estimation of investment and operation costs in comparison to other sensor types is given.
- Design of low-cost, compact and weather-proof whole sky imagers for high-dynamic-range captures45
Two new Wide Angle High-Resolution Sky Imaging System (WAHRSIS) models are presented, which were designed especially to withstand the hot and humid climate of Singapore.
- Visibility of noise in natural images45
This work investigates the visibility of noise itself as a target and uses natural images as the masker and conducts psychophysical experiments to determine the detection threshold of these noise targets on many different types of image content.
- Regional T2 and Sodium Concentration Estimates in the Normal Human Brain by Sodium-23 MR Imaging at 1.5 T45
It is found that regional T2 measurements more consistently distinguish the high signal regions of cortex, CSF, and blood than do regional sodium estimates, whereas regional Sodium estimates adequately distinguish white matter from cortex.
- Implicit User-centric Personality Recognition Based on Physiological Responses to Emotional Videos40
Personality differences are better revealed while analyzing responses to emotionally homogeneous responses to affective movie clips, and significantly above-chance recognition is achieved for all five traits.
- From microscope to micropixels: A rapid review of artificial intelligence for the peripheral blood film39
Despite the high accuracy, the challenges toward real world translational usage of these AI trained models include the need for well-validated multicentre data, data standardisation, and studies on less common cell types and non-malarial blood-borne parasites.
- Empirical Analysis Of Overfitting And Mode Drop In Gan Training39
It is shown that when stochasticity is removed from the training procedure, GANs can overfit and exhibit almost no mode drop, providing evidence against prevailing intuitions that GAns do not memorize the training set, and that mode dropping is mainly due to properties of the GAN objective rather than how it is optimized during training.
- Recognition of Advertisement Emotions With Application to Computational Advertising39
An affective video ad dataset is compiled which evokes coherent emotions across users, and the efficacy of content-centric convolutional neural network features for ad AR vis-ã-vis handcrafted audio-visual descriptors is explored.
- 39
- Emotion-based sequence of family photos38
This paper presents a method for the automatic creation of slideshows from family photo collections based on the emotions of a given group of people that makes use of a new image dissimilarity function, which can integrate various attribute combinations and preferences, making the system highly user-adaptable and flexible.
- Toward perceptual metrics for video watermark evaluation38
This paper proposes two metrics for evaluating the visual impact of video watermarks based on several different watermarking algorithms and video sequences, and identifies the most prominent impairments as spatial noise and temporal flicker.
- Vision-Based Projected Tabletop Interface for Finger Interactions37
A vision-based projected table-top interface for finger interaction that provides more comfortable and direct viewing for users, and more natural, intuitive yet flexible interaction than classical or tangible interfaces.
- 36
- Xenon and Krypton as Radiographic Inhalation Contrast Media With Computerized Tomography: Preliminary Note36
Computerized tomography scans of samples of the inert gases xenon and krypton inside a Lucite phantom, show these gases to be potentially useful inhalation contrast media with CT scanning of the chest.
- Mimic-IV-ICD: A new benchmark for eXtreme MultiLabel Classification34
This paper proposes a public benchmark suite for ICD-10 coding using a large EHR dataset derived from MIMIC-IV, the most recent public EHR datasets, and implements and compares several popular methods for I CD coding prediction tasks to standardize data preprocessing and establish a comprehensive ICD coding benchmark dataset.
- MAXIMIZING AUDIOVISUAL QUALITY AT LOW BITRATES34
Using thesedata, the insuence of video codecs and framerate as well as audio channels and sampling rate on qual-ity is explored, and the optimal trade-of f between bits allocated toaudio and video inside a bitstream is investigated.
- Multi-label Cloud Segmentation Using a Deep Network33
This paper proposes to use a deep learning architecture (U-Net) to perform multi-label sky/cloud image segmentation and outperforms recent literature by a large margin.
- Stereo/multiview picture quality: Overview and recent advances32
This paper reviews stereo/multiview picture quality from an engineering perspective with a focus on recent or emerging approaches and technologies used in 3D systems, addressing in particular depth issues, multiview issues, display issues, viewer issues, as well as possible measurements and standards for 3D quality.
- Estimation of solar irradiance using ground-based whole sky imagers31
This paper shows how images taken by WSIs can be used to estimate solar radiation, and includes ground-based weather stations at the same location as the imagers to validate the methods.
- Quantification of pollution loads from CSOs into surface water bodies by means of online techniques30
A modular online monitoring station suitable for continuous application in sewer networks, wastewater treatment plants and receiving water bodies has been designed and installed and serves as the backbone of a water quality pilot network.
- Forecasting Precipitable Water Vapor Using LSTMs29
Long-Short-Term-Memory networks have been used extensively for time series forecasting in recent years due to their ability of learning patterns over different periods of time, and this ability is applied to learning the pattern of Global Positioning System-based Precipitable Water Vapor measurements over a period of 4 hours.
- High-dynamic-range imaging for cloud segmentation28
This proposed approach is the first using HDR radiance maps for cloud segmentation using High-Dynamic-Range (HDR) imaging based on multi-exposure fusion and achieves very good results.
- Inferring painting style with multi-task dictionary learning28
This paper presents a novel dictionary learning approach to automatically uncover the artistic style from paintings, by automatically decoupling style-specific and artist-specific patterns, and shows that this approach significantly outperforms state-of-the-art methods.
- 28
- Perceptual quality assessment for video watermarking27
This work proposes a methodology for evaluating the visual quality of watermarked video using a software tool that measures different types of perceptual video artifacts and designs the corresponding objective metrics.
- 26
- 24
- EEG-based Evaluation of Cognitive Workload Induced by Acoustic Parameters for Data Sonification23
This work is the first to automatically estimate the cognitive load induced by different acoustic parameters considered for sonification, implying a congruent relationship between explicit and implicit user responses.
- Rough-Set-Based Color Channel Selection23
This letter proposes the use of rough sets for color channel selection in visible-light images by assessing color channels with respect to their contribution for segmentation and identifies the most effective ones.
- PET: An eye-tracking dataset for animal-centric Pascal object classes23
The authors' database comprises eye movement recordings compiled from forty users for the bird, cat, cow, dog, horse and sheep trainval sets from the VOC 2012 image set and shows how feature pooling around fixated locations enables enhanced (animal) object classification accuracy.
- 23
- 22
- 22
- Short-term prediction of localized cloud motion using ground-based sky imagers22
Hemispherical images of the sky at regular intervals of time using ground-based cameras provide a high resolution and localized cloud images, and two successive frames are used to compute optical flow and predict the future location of clouds.
- 22
- Benefits, limitations and uncertainty of in situ spectrometry22
In order to assess the total uncertainty of in situ spectrometry, a comprehensive lab test was carried out, which includes a detailed investigation of the uncertainty of CODreference methods (DIN and small tube tests), a precision analysis of the absorption spectra and the impact of those two factors on thetotal uncertainty of the COD(EQ) correlation model.
- Artificial Intelligence-Generated Facial Images for Medical Education21
This work evaluated the use of text-to-image models (Microsoft’s Bing Image creator and Shutterstock’s AI image generator) to generate realistic images of human faces and their associated pathology to augment rare medical image datasets for medical education.
- 21
- MR image segmentation using vector decomposition and probability techniques: A general model and its application to dual‐echo images21
A general model is developed for segmenting magnetic resonance images using vector decomposition and probabilfty techniques, and can produce com‐posrte tissue‐type images similar to that of the probability methods, by comparing the fractional volumes assigned to different tissues on each voxel.
- Tangible mixed reality desktop for digital media management20
A tangible mixed reality desktop that supports gesture-oriented interactions in 3D space based on computer vision techniques for hand and finger detection, without the need for attaching any devices to the user.
- 20
- Dual‐echo MRI segmentation using vector decomposition and probability techniques: A two‐tissue model20
A vector decomposition technique with Gaussian probability thresholding in feature space is combined to segment normal brain tissues, tumors, or other abnormalities on dual‐echo MR images to minimize contamination from other tissues.
- Identifying psychiatric manifestations in schizophrenia and depression from audio-visual behavioural indicators through a machine-learning approach19
The results suggest that machine-learning models leveraging audio-visual characteristics can help diagnose, assess, and monitor patients with schizophrenia and depression.
- 19
- 19
- 19
- 17
- Episodic CAMN: Contextual Attention-Based Memory Networks with Iterative Feedback for Scene Labeling17
- Study of clear sky models for Singapore17
This paper performs a comparative analysis of several popular clear-sky models, in the tropical region of Singapore, and concludes the most reliable clear sky model for Singapore, based on all clear sky days in a year.
- 17
- 17
- 17
- 16
- 16
- 16
- 16
- Wavelet-based contrast computation and application to digital image watermarking16
An isotopic measure of local contrast for natural images that is based on analytic filters is introduced and the design of directional wavelet frames suitable for its computation is presented.
- Detecting rainfall onset using Sky Images15
Ground-based sky cameras are used to detect the onset of rainfall and these images contain additional information about cloud coverage and movement and are therefore useful for accurate rainfall nowcast.
- Efficient measurement of stereoscopic 3D video content issues15
These metrics are designed for high computational efficiency to permit real-time video content analysis, and are shown to be robust to common video impairments.
- 15
- Respiratory Depression in Goats by Stable Xenon15
In a study in goats, it is demonstrated that xenon has a characteristic respiratory depressant effect, unlike effects of comparable anesthetic concentrations of nitrous oxide and halothane.
- Trusted Media Challenge Dataset and User Study14
The results show that the TMC dataset can fool human participants in many cases, and is released to enable further research.
- 14
- A Chaotic Approach on Solar Irradiance Forecasting13
This study reveals that the optimum dimension required for solar irradiance measurement is 4 for both samplings, indicating that a minimum of 4 dimensions is required for embedding the data for the best representation of input.
- 13
- Precision pointing H∞ control design for absolute, window-, and stability-time errors13
This article extends the standard multi-objective H2/H∞ control problem to explicitly take into account requirements on pointing error index performance, and develops a derivation of a control design approach that subjects the closed-loop control system to only one matrix criterion, the H∞-norm.
- 13
- PULSAR: Pre-training with Extracted Healthcare Terms for Summarising Patients' Problems and Data Augmentation with Black-box Large Language Models12
The proposed approach to this task, which integrates two complementary components, employs large language models (LLMs) for data augmentation; the other is an abstractive summarisation LLM with a novel pre-training objective for generating the patients’ problems summarised as a list.
- 12
- Information Flow Between Requirement Artifacts. Results of an Empirical Study12
The results of an empirical study of artifacts created and used in the requirements engineering process are presented, finding that meeting notes and lists of requirements are most commonly used, that specification documents are information sinks, and that most projects create several different artifacts.
- Supply Chain Event Management With Software Agents12
A concept based on software agent technology is presented which enhances time and defect flexibility of supply chain processes and indicates cost and cycle time reductions in multi-level supply chains which are not achieved by conventional approaches.
- 12
- 11
- Multimodal Study of the Effects of Varying Task Load Utilizing EEG, GSR and Eye-Tracking11
This study suggests that while EEG is a good predictor of task performance, additional modalities such as GSR increase the likelihood of more accurate predictions.
- QoS provisioning by power control for video communication via satellite links11
Simulation results show the effectiveness of the proposed power control solution and its ability to sustain video quality levels in spite of rain fades.
- User Studies of a Multiplayer First Person Shooting Game with Tangible and Physical Interaction11
This system provides an intuitive way for the users to interact with the virtual world by physically moving around the real world and aiming freely with tangible objects to encourage physical interaction between the players as they compete or collaborate with each other.
- 11
- 10
- Interactive Multi-camera Soccer Video Analysis System10
An end-to-end system for multi-camera soccer video analysis that makes heavy use of parallel processing for optimization of the processing workflow and the proposed thread-level parallelism speeds up the system by more than 15 times while maintaining the level of accuracy.
- A general framework for image feature matching without geometric constraints10
It is proved that the traditional Ratio-Match is the worst performer, and a general framework of 4 algorithms for image feature matching is proposed, proving that all three methods consistently outperform or equal ratio-Match.
- Brush-and-drag10
An effective and easy-to-use brush-and-drag interface that allows the user to interactively explore and compare photos within a broader scene context and demonstrates to offer effective interactions in a user study.
- 10
- 10
- Open source OCR framework using mobile devices10
This work developed a complete OCR framework with subsystems from open source desktop community, including a popular open source OCR engine named Tesseract for text detection & recognition and Flite speech synthesis module, for adding text-to-speech ability.
- A Tangible Game Interface Using Projector-Camera Systems10
A tangible game interface using projector-camera systems that provides more direct viewing as well as more natural and flexible interaction than bulky HMD's or monitor-based game interfaces.
- 3D surveillance system using multiple cameras10
This application is concerned with identifying humans in the scene and then identifying their postures, which can help with automatic threat assessment of a scene.
- Representation Learning of Structured Data for Medical Foundation Models9
The UniStruct architecture is introduced to design a multimodal medical foundation model of unstructured text and structured data, which addresses the challenges LLMs face in processing medical codes due to the shortcomings of current tokenization methods.
- 9
- Cloud radiative effect study using sky camera9
It is concluded that higher cloud coverage greatly impacts the total solar irradiance, and such studies will immensely help in solar energy generation and forecasting.
- 9
- Modeling Image Appeal Based on Crowd Preferences for Automated Person-Centric Collage Creation9
This paper employed nine low level image attributes to model the image selection process, and trained SVRs which could adequately predict image selections for the album-specific conditions, suggesting that context greatly influences the categorization of what is more and less appealing.
- 9
- 9
- In-line river monitoring – new challenges and opportunities9
The operational problems and experiences of collecting data over a period of one year in the Danube River downstream of Vienna are described, which shows that the values are reliable and therefore applicable to further interpretations.
- A Feature Map Approach to Real-Time 3-D Object Pose Estimation from Single 2-D Perspective Views9
A novel approach to the computation of an approximate estimate of spatial object pose from camera images is proposed based on a neural network that generates pose hypotheses in real time, which can be refined by registration or tracking systems.
- Spontaneous Hemorrhage into an Extralobar Bronchopulmonary Sequestration9
In a patient with a previously normal chest film, the appearance of a smooth mass lesion in the paravertebral area, particularly when adjacent to the diaphragm, should raise the possibility of this entity.
- 8
- Morphset: Augmenting Categorical Emotion Datasets With Dimensional Affect Labels Using Face Morphing8
This work proposes a method to generate synthetic images from existing categorical emotion datasets using face morphing as well as dimensional labels in the circumplex space with full control over the resulting sample distribution, while achieving augmentation factors of at least 20x or more.
- 8
- Special issue on Image and Video Quality Assessment8
An overview of state-of-the-art IQA/VQA methods is provided and new developments towards various directions are addressed, including a novel image quality metric based on gradient profiles and the other on structural similarity.
- Upgrading of wastewater treatment plants for nutrient removal under optimal use of existing structures8
The two stage activated sludge HYBRID-process often provides an interesting design alternative for such plant upgrades, especially in case an anaerobic sludge treatment stage already exists.
- A feature map approach to pose estimation based on quaternions8
The rigid map is introduced, which is derived from Kohonen's self-organizing feature map, and its topology is fixed and chosen in accordance with the quaternion representation.
- Clinical and Experimental Sodium Magnetic Resonance Imaging8
In conclusion, sodium MR imaging has potential for providing physiologic information relevant to cell mitosis, cell energy state, rCBV, and seizures and should become a routine examination in the clinical setting.
- A study of image discrepancies due to object time-dependence in transmission and emission tomography8
The authors have studied the properties of such artefacts by the computer simulation of decaying exponential time-dependence in three different spatial distributions and four transmission and emission CT geometries, and found that artefacts are reduced for all geometry as scan duration is reduced.
- 7
- Intuitive user interface for mobile devices based on visual motion detection7
This paper proposes intuitive motion-based UI controls for mobile devices with built-in cameras based on the visual detection of the device's self-motion that is well received by the users and clearly preferred over traditional button-based controls.
- 7
- Intuitive Map Navigation on Mobile Devices7
Results show that users appreciate the motion-based user interface and find it more intuitive than traditional key-based controls, even though there is a learning curve.
- 7
- Vision-Based Autonomous Landing for Small UAVs - First Experimental Results7
This paper describes the development of intelligent vision capabilities for small sized UAVs (Unmanned Aerial Vehicles) as part of the program “CAROLO”, currently underway at the Technische Universitat Braunschweig, Germany, to create image processing algorithms and appropriate hardware that suites the requirements of small sizedUAVs.
- 6
- Efficient Training of Self-Supervised Speech Foundation Models on a Compute Budget6
This paper investigates how to efficiently train speech foundation models with self-supervised learning (SSL) under a limited compute budget and identifies a trade-off between model size and data size, highlighting an optimal model size for a given compute budget.
- Artificial intelligence‐generated peripheral blood film images using generative adversarial networks and diffusion models6
This paper presents a meta-modelling architecture that automates the very laborious and time-intensive and therefore expensive and expensive and therefore time-heavy and expensive process of manually cataloging and decrypting data.
- 6
- 6
- 6
- Investigating the generalizability of EEG-based cognitive load estimation across visualizations6
Experiments reveal that CL estimation suffers across visualizations calling for for effective machine learning techniques to benchmark visual interface usability for a given analytic task.
- PhotoCluster: A Multi-clustering Technique For Near-duplicate Detection In Personal Photo Collections6
This paper presents PhotoCluster, a new technique for identifying non-identical near-duplicate images in personal photo collections that produces a non-binary near-double probability for each image pair, which exhibits correlation with the average observer opinion.
- 6
- Jointly Estimating Interactions and Head, Body Pose of Interactors from Distant Social Scenes6
This work exploits positional and pose cues characterizing interactors and interactions to jointly infer F-formations and head, body pose of interactors in a social scene captured by surveillance cameras.
- X-Eye: A reference format for eye tracking data to facilitate analyses across databases6
This paper proposes a common reference format for eye tracking data, together with conversion routines for 16 existing image eye tracking databases to that format, and conducts a few analyses on these datasets as examples of what X-Eye facilitates.
- 6
- Domain Predictive Control Under Uncertain Numerical State Information6
This work shows how to propagate and reason about uncertain state information directly in the planning process, enabling hybrid systems to find plans that satisfy numerical goals with predefined confidence.
- Investigative monitoring in the context of detecting anthropogenic impact on an epipotamal river6
The analysis of the online data clearly showed that some of the measured parameters can be used as indicators for specific discharges, which from additional targeted investigations could be identified as contributors to the foaming problems.
- 6
- A Comparison of MEMS-Based IMU Calibration Models for Autonomous MAV Navigation6
First and second order approximations for the calibration tables turn out to provide equivalent navigation accuracy compared to the original calibration tables by saving computational power.
- Development of a Fully Autonomous Micro Aerial Vehicle (MAV) for Ground Traffic Surveillance6
The currect prototype “Carolo”, an autonomously operating Micro Aerial Vehicle with a wingspan of 40 cm and a mass of 390 g, is introduced, one of the main applications for the MAV is ground traffic surveillance, allowing quick and flexible visual information retrieval in real-time.
- 5
- Detecting Blurred Ground-based Sky/Cloud Images5
This approach is the first of its kind in the automatic identification of blurred images for ground-based sky/cloud images, using a static external marker and has a detection accuracy of 94%.
- 5
- Give Me One Portrait Image, I Will Tell You Your Emotion and Personality5
An end-to-end trainable and deep Siamese-like network that can take one portrait photo as input and predict one's Big-Five apparent personality as well as emotion attributes and demonstrates the feasibility of inferring the apparent personality directly fro emotion.
- 5
- Measuring Neural, Physiological and Behavioral Effects of Frustration5
Electroencephalography (EEG), galvanic skin response (GSR), and gaze tracking data, and subjective data from a NASA Task-Load-Index based questionnaire are collected to assess frustration during task performance.
- Model-based support for Product-Service System planning5
The presented planning model and its implementation in a visualization tool help the Product-Service System (PSS) planner to maintain an overview of the system, the relations between its elements and further necessary information.
- 5
- 5
- Impairment metrics for MC/DPCM/DCT encoded digital video5
LTS1 Reference LTS-CONF-2001-046 Record created on 2006-06-14, modified on 2016-08-08.
- 4
- 4
- Subjective Quality Assessment of Ground-based Camera Images4
A new image quality assessment dataset containing original and distorted nighttime images of sky/cloud from SWINSEG database is presented and statistical analyses of the subjective scores showed the impact of noise type and distortion level on the perceived quality.
- 4
- Articulated Object Registration Using Simulated Physical Force/Moment for 3D Human Motion Tracking4
This paper presents a 3D registration algorithm based on simulated physical force/moment for articulated human motion tracking and introduces a hierarchical scheme for model state updating, which automatically incorporates human kinematic constraints.
- 4
- 4
- 4
- 3
- Artificial intelligence generated leukemia cell images3
This paper presents a meta-modelling architecture that automates the very laborious and time-intensive and therefore expensive and expensive and therefore time-heavy and expensive process of manually cataloging and decrypting data.
- 3
- 3
- Sparse Code Filtering for Action Pattern Mining3
This paper proposes a sparse code filtering (SCF) framework which can mine the action patterns of multi-view action recognition by integrating the classifiers and the class-wise sparse coding process into a collaborative filtering (CF) framework to mine the discriminative sparse codes and classifiers jointly.
- 3
- 3
- 3
- Recovering badly exposed objects from digital photos using internet images3
This paper considers the problem of clipped-pixel recovery over an entire badly exposed image region, using two correctly exposed images of the scene that may be captured under different conditions, and uses two reference images to recover texture and color.
- How do users make a people-centric slideshow?3
Results show that there is correlation between specific predictors and selected images, as well as their relative position in the final sequence, which indicates that a crowdsourcing experiment will indeed highlight the characteristics of the average user, which can be incorporated into an automatic people-centric slideshow creator.
- The Rain Attenuation on Real-Time Video Streaming via Satellite Links3
The real-time video transmission via satellite links is discussed and the rainfall attenuation on the quality of video is analyzed to assess the impact of rain on the video quality.
- 3
- 3
- 3
- 3
- Affordable Real Time Cartography Using the MAV CAROLO - Limitations and Prospects3
The developed simulation tool box was used to study the effects different attitude estimation algorithms have on the quality of the produced image material and the biggest challenge proves to be attitude estimation for the MAV.
- Treatment of easily biodegradable wastewater avoiding bulking sludge3
A new process that has been developed for a dairy in Austria that is completely free from biodegradable material and can without danger of bulking be fed to the aeration tank.
- Seemingly Plausible Distractors in Multi-Hop Reasoning: Are Large Language Models Attentive Readers?2
- 2
- A center-surround framework for spatial image processing2
A computational framework inspired by the center-surround antagonistic receptive fields of the human visual system is presented that can give rise to existing spatial image processing techniques, including adaptive tone-mapping, local contrast enhancement, text binarization and local feature detection.
- 2
- 2
- 2
- What impacts skin color in digital photos?2
This work employs a statistical framework to automatically build color models from image datasets crawled from the web and assess both technical and semantic metadata and shows that semantic metadata has a more significant impact, which suggests that semanticmetadata holds important cues for processing of skin colors.
- Perceptual Semantics2
This chapter prioritizes the visual data in order to improve the compression performance of video coders and the prediction performance of perceptual quality metrics, and proposes a semantic segmentation stage, which takes into account some aspects of the cognitive behavior of people when watching a video.
- An efficient stochastic framework for 3D human motion tracking2
The proposed stochastic framework for articulated 3D human motion tracking uses explicit 3D reconstructions of the human body based on a visual hull algorithm as its system input, which effectively eliminates self-occlusion.
- 2
- The Identification of Pneumoconiosis From Hospital Discharge Data2
The use of hospital discharge summaries by Valiente and Rosenman 1 to identify cases of silicosis recalls a preliminary study carried out in 1987, where the yield seemed to be less than the expected incidence to Dr S. S. Winkler.
- 2
- 1
- Evaluation of a Semi‐Supervised AI Model ( ASUS Blade) for Peripheral Blood Film Leukocyte Classification1
Blade (a semi‐supervised AI model) and CellaVision DM9600 (commercial benchmark) are evaluated against MM in automated leukocyte classification in peripheral blood film analysis.
- 1
- 1
- Generating near-infrared facial expression datasets with dimensional affect labels1
Two complementary data augmentation methods are presented to create NIR image datasets with dimensional emotion labels from existing categorical and/or visible-light datasets, showing that these generated NIR datasets are comparable to existing datasets in terms of data quality and baseline prediction performance.
- 1
- 1
- 1
- 1
- System for creating slideshows based on people and their emotions1
The intuitive interface allows the user to provide a personalized definition of image similarity, resulting in different image sequences, and making the system highly user-adaptable and flexible.
- 1
- Observability of Star Tracker / Gyro Based Attitude Estimation Considering Time-Variant Sensor Misalignment1
The fusion of measurements from star trackers and gyroscopes within optimal estimators/filters is a common approach for spacecraft attitude determination, where the focus is on deterministic time-variant misalignment between star tracker and gyroscope unit.
- Performance analysis of visual tracking algorithms for motion-based user interfaces on mobile devices1
This paper compares the performances of three feature or landmark-guided motion tracking algorithms, namely marker-based tracking with MXRToolkit, face tracking based on CamShift, and MonoSLAM and analyzes the robustness of the algorithms by testing for manipulative factors like image scaling and frame-skipping.
- Intuitive application-specific user interfaces for mobile devices1
This paper proposes intuitive user interfaces for two applications on mobile phones with built-in cameras, namely, a car-racing game and a map navigation application based on the visual detection of the device's self-motion.
- 1
- A post-processing technique for displaying vessels from routine fast-spin-echo images: MRI-derived angiography1
By applying an enhancement filter to a stack of routine fast-spin-echo MR images, projected angiographic images can be generated that provide additional vessel information that is useful when MR angiography is unavailable or as an aid in planning dedicated MRAngiographic studies.
- 1
- 1
- 1
- 1
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- ADGD'21–
ADGD21: 1st Workshop on Synthetic Multimedia - Audiovisual Deepfake Generation and Detection is organized so as to provide a platform for researchers and engineers to share their ideas and approaches in this field.
- –
- –
- Integration of Scenario-based Requirements Forecast into Model-based Product-service System Planning–
- –
- –
- –
- A WEB-BASED INTRODUCTION TO BIOINFORMATICS - Epistemological Beliefs in Bioinformatics–
An OpenLab learning course was used to reflect the students` and undergrads` impression of bioinformatics and found the increase of knowledge is significantly higher when doing additional data analysis and computer work than working only in the wet lab.
- –
- Welcome message from the QoMEX 2013 chairs–
It is the authors' pleasure to welcome you to QoMEX 2013, the Fifth International Workshop on Quality of Multimedia Experience, in Klagenfurt am Worthersee, Austria.
- –
- –
- A WEB-BASED INTRODUCTION TO BIOINFORMATICS - Epistemological Beliefs in Bioinformatics–
An OpenLab learning course was used to reflect the students` and undergrads` impression of bioinformatics and found the increase of knowledge is significantly higher when doing additional data analysis and computer work than working only in the wet lab.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- A Simple Method of Obtaining Single-frame Enlargements of Cine Film–
In cine studies of the gastrointestinal tract, the cardiovascular system, and the urinary tract a single frame often best shows the fistula, the intravascular shunt, or the urethral valve and a simple method of enlarging and preserving this single frame at the time the cine film is projected is of great value for teaching and easy reference.
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 2026-10-12. 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.