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
- Gian Luca ForestiSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works209 from public data
- 539
- 496
- The Eighth Visual Object Tracking VOT2020 Challenge Results350
A significant novelty is introduction of a new VOT short-term tracking evaluation methodology, and introduction of segmentation ground truth in the VOT-ST2020 challenge – bounding boxes will no longer be used in theVDT challenges.
- Wide-Slice Residual Networks for Food Recognition228
This work introduces a new deep scheme that is designed to handle the food structure, and first introduces a slice convolution block to capture the vertical food traits that are common to a large number of categories.
- NTIRE 2020 Challenge on Real-World Image Super-Resolution: Methods and Results191
The NTIRE 2020 challenge addresses the real world setting, where paired true high and low-resolution images are unavailable, and the ultimate goal is to achieve the best perceptual quality, evaluated using a human study.
- Saliency Weighted Features for Person Re-identification144
This work proposes a novel person re-identification approach, inspired by human gazing capabilities, that uses the salient regions of a given person as a weighting tool in the image feature extraction process and is compared with best state-of-the-art approaches to show its overall superior performance.
- 140
- 130
- The Ninth Visual Object Tracking VOT2021 Challenge Results127
The Visual Object Tracking challenge VOT2021 is the ninth annual tracker benchmarking activity organized by the VOT initiative; results of 71 trackers are presented; many are state-of-the-art trackers published at major computer vision conferences or in journals in recent years.
- Temporal Model Adaptation for Person Re-identification123
This work introduces a similarity-dissimilarity learning method which can be trained in an incremental fashion by means of a stochastic alternating directions methods of multipliers optimization procedure and exploits a graph-based approach to present the user only the most informative probe-gallery matches that should be used to update the model.
- 119
- 114
- 109
- 106
- 105
- AIM 2020 Challenge on Efficient Super-Resolution: Methods and Results101
The AIM 2020 challenge on efficient single image super-resolution was to super-resolve an input image with a magnification factor x4 based on a set of prior examples of low and corresponding high resolution images with focus on the proposed solutions and results.
- 95
- 95
- NTIRE 2024 Challenge on Image Super-Resolution (×4): Methods and Results83
This paper reviews the NTIRE 2024 challenge on image super-resolution (×4), highlighting the solutions proposed and the outcomes obtained, and offers a comprehensive overview of current trends in this field.
- A network of co-operative cameras for visual surveillance80
A network of co-operative cameras for the visual surveillance of parking lots employs multiple subnets able to manage static and active cameras in a hierarchical framework and is able to track multiple targets simultaneously and in real-time throughout the controlled areas.
- Siam-U-Net: encoder-decoder siamese network for knee cartilage tracking in ultrasound images79
This study proposes a new deep learning method to track, accurately and efficiently, the femoral condyle cartilage in ultrasound sequences, which were acquired under several clinical conditions, mimicking realistic surgical setups.
- 73
- 72
- 69
- 69
- 63
- Deep Generative Adversarial Residual Convolutional Networks for Real-World Super-Resolution61
A deep Super-Resolution Residual Convolutional Generative Adversarial Network (SRResCGAN1) to follow the real-world degradation settings by adversarial training the model with pixel-wise supervision in the HR domain from its generated LR counterpart.
- Aggregating Deep Pyramidal Representations for Person Re-Identification60
It is argued that a pyramid-inspired method capturing multi-scale information may overcome requirements of extensive human annotation labor or demand additional computational efforts to overcome the challenges of triplet-similarity learning.
- A supervised extreme learning committee for food recognition60
Experimental results show that the proposed food recognition system outperforms state-of-the-art works on four publicly available benchmark datasets.
- Stereo rectification of uncalibrated and heterogeneous images59
An algorithm for rectifying heterogeneous and uncalibrated pairs of stereo images is presented and improvements of the proposed method over direct rectification over nonlinear constrained optimization are assessed.
- Visual Object Tracking in First Person Vision51
It is proved that trackers bring benefits to FPV downstream tasks requiring short-term object tracking and expected that generic object tracking will gain popularity in FPV as new and FPV-specific methodologies are investigated.
- 49
- Occlusion-aware multiple camera reconfiguration48
The proposed technique works by approximating the observed zones with ellipses and finds a locally optimal solution by using the Expectation Maximization algorithm.
- 46
- 45
- NTIRE 2021 Challenge on Burst Super-Resolution: Methods and Results43
The NTIRE2021 challenge on burst super-resolution was to generate a clean RGB image with 4 times higher resolution given a RAW noisy burst as input, and the top-performing methods set a new state-of-the-art for the burstsuper-resolution task.
- AIM 2020 Challenge on Real Image Super-Resolution: Methods and Results43
This paper introduces the real image Super-Resolution challenge that was part of the Advances in Image Manipulation (AIM) workshop, held in conjunction with ECCV 2020, and gauges the state-of-the-art approaches for real image SR in terms of PSNR and SSIM.
- 41
- Coverage Control of Multiple Unmanned Aerial Vehicles: A Short Review39
A multiple unmanned aerial vehicle (MUAV) system is a group of vehicles that are designed with the aim to perform some collective behavior.
- 39
- 37
- Detecting moving people in video streams37
A finite state automaton has been designed to give a robust classification and to reduce the number of false or missed blobs and a bounding ellipse is computed for each detected blob to isolate moving people.
- A balanced neural tree for pattern classification36
A new neural tree architecture, balanced neural tree (BNT), is proposed to reduce tree size and improve classification with respect to classical NTs to achieve this result through perceptron substitution and pattern removal.
- 36
- 36
- 35
- Generalized neural trees for pattern classification35
Its particular structure provides an easily probabilistic interpretation of the pattern classification task and allows growing small neural trees with good generalization properties and displays good classification performances for training sets having complex distributions.
- A Structured Committee for Food Recognition34
A committee-based recognition system that chooses the optimal features out of the existing plethora of available ones that outperforms state-of-the-art works on the most used three publicly available benchmark datasets.
- 34
- Bituminous Mixtures Experimental Data Modeling Using a Hyperparameters-Optimized Machine Learning Approach33
This study introduces a machine learning approach based on Artificial Neural Networks (ANNs) for the prediction of Marshall test results, stiffness modulus and air voids data of different bituminous mixtures for road pavements.
- Visual Tracking by Means of Deep Reinforcement Learning and an Expert Demonstrator33
A3CT, which exploits demonstrations of a state-of-the-art tracker to learn an effective tracking policy, and A3CTD, that takes advantage of the same expert tracker to correct its behaviour during tracking are proposed.
- Modeling feature distances by orientation driven classifiers for person re-identification32
A method to extract multiple frames of the same person with different orientations in order to capture the complete person appearance and to learn the pairwise feature dissimilarities space (PFDS) formed by the subspaces of similar and different image pair orientations.
- Real-time image processing for active monitoring of wide areas31
A real-time active system able to monitor wide areas by processing video sequences acquired by a pan, tilt, and zoom (PTZ) camera is presented and a new feature clustering method is studied to reject badly tracked features.
- 30
- 28
- Is First Person Vision Challenging for Object Tracking?27
The study extensively analyses the performance of recent visual trackers and baseline FPV trackers with respect to different aspects and considering a new performance measure, and shows that object tracking in FPV is challenging.
- Is First Person Vision Challenging for Object Tracking? The TREK-100 Benchmark Dataset.27
The study extensively analyses the performance of recent visual trackers and baseline FPV trackers with respect to different aspects and considering a new performance measure, and shows that object tracking in FPV is challenging.
- 26
- A cooperative multicamera system for video-surveillance of parking lots26
A cooperative system able to manage static and active cameras in a hierarchical framework is presented and is able to track multiple targets simultaneously and in real-time through the controlled areas.
- Real-Time Incremental and Geo-Referenced Mosaicking by Small-Scale UAVs25
This paper proposes a small-scale UAV based system for real-time creation of incremental and geo-referenced mosaics of video streams acquired at low-altitude and presents several innovative contributions, including the use of A-KAZE feature extractor in aerial images, a Region Of Interest (ROI) to speed-up the stitching stage.
- 25
- 24
- 23
- Sparse Matching of Random Patches for Person Re-Identification23
A novel algorithm is introduced that randomly samples a set of image patches to compute the person signature and uses the correlation matrix computed between such patches as a weighing tool in the signature matching process.
- Stereo Localization Using Dual PTZ Cameras23
The proposed stereo localization system has two advantages: improved localization on a partially occluded target and monitoring a large environment using only two PTZ cameras without missing significant information.
- 22
- A pool of multiple person re-identification experts22
A Multiple Re-identification Experts (MuRE) are trained to reliably match new probes and the answers from all the experts are combined to achieve a final decision, inspired by the recent police lineup innovations.
- 22
- 22
- An Intelligent Auto-Organizing Aerial Robotic Sensor Network System for Urban Surveillance20
This is the first proof-of-concept of a fully integrated intelligent auto-organizing aerial robotic sensor network system of cameras installed on UAVs and ground fixtures for urban surveillance using the decentralised control paradigm.
- Deep Cyclic Generative Adversarial Residual Convolutional Networks for Real Image Super-Resolution19
This work proposes the Super-Resolution Residual Cyclic Generative Adversarial Network (SRResCycGAN) by training with a generative adversarial network (GAN) framework for the LR to HR domain translation in an end-to-end manner and demonstrates that the proposed approach achieves comparable results as the other state-of-art methods.
- 19
- Person Re-Identification via Unsupervised Transfer of Learned Visual Representations18
A deep learning scheme which leverages on the large quantity of labeled data that may be available in a source domain to learn a robust visual representation is proposed which is exploited in an unsupervised transfer learning scheme to better handle the difficulties in the target re-identification domain.
- 18
- A neural tree for classification using convex objective function17
A neural tree based classifier, called the convex objective function neural tree (COF-NT), which has a specialized perceptron at each node which calculates the errors before the neuron's non-linear activation function instead of after them and is able to reduce the training time without decreasing classification accuracy.
- Audio–video biometric recognition for non-collaborative access granting17
An optimization of a speech de-noising algorithm to optimize the performance of the extended Kalman filter (EKF) and a new temporal classification is proposed to improve the robustness of the frame-by-frame classification.
- 16
- Outdoor Environment Monitoring with Unmanned Aerial Vehicles16
Experimental results prove that the proposed approach can be used to detect relevant changes in the specific case of road safety assurance in dangerous zones.
- Visual Sensor Technology for Advanced Surveillance Systems: Historical View, Technological Aspects and Research Activities in Italy16
The paper describes the main characteristics of an advanced visual sensor network that directly processes locally acquired digital data, automatically modifies intrinsic and extrinsic parameters to increase the quality of acquired data and automatically selects the best subset of sensors in order to monitor a given moving object in the observed environment.
- 16
- Visualizing Skiers' Trajectories in Monocular Videos15
SkiTraVis is proposed, an algorithm to visualize the sequence of points traversed by a skier during its performance and demonstrates the potential of the solution for broadcasting media enhancement and coach assistance.
- Multi Branch Siamese Network For Person Re-Identification15
A multi branch Siamese Deep Neural Network with multiple classifiers to overcome the above issues and creates a stronger descriptor with fine-grained information from global features of a person.
- 15
- Tracking Skiers from the Top to the Bottom14
The largest and most annotated dataset for computer vision in skiing, SkiTB, is introduced and several visual object tracking algorithms are tested, providing valuable insights into the applicability of different tracking methods for vision-based skiing analysis.
- Deep interactive encoding with capsule networks for image classification14
Residual Capsule Network (ResNetCaps) and Dense Capsule network (DenseNetC Caps) are proposed to introduce to tackle the image recognition problem and the application of feature interaction methods between capsules to promote their cooperation while dealing with complex data is investigated.
- 14
- Online Episodic Memory Visual Query Localization with Egocentric Streaming Object Memory13
This work introduces Online Visual Query 2D (OVQ2D), a task where models process video streams online, observing each frame only once, and retrieve object localizations using a compact memory instead of full video history, and addresses ESOM (Egocentric Streaming Object Memory), a novel framework integrating an object discovery module, an object tracking module, and a memory module that find, track, and store spatiotemporal object information for efficient querying.
- 13
- 13
- Deep convolutional feature details for better knee disorder diagnoses in magnetic resonance images13
The experimental results achieved demonstrate that the application of MRPyrNet to baseline methodologies improves their diagnostic capability, especially in the case of anterior cruciate ligament tear and meniscal tear because of MRpyrNet's ability in exploiting the relevant appearance features of such disorders.
- CoCoLoT: Combining Complementary Trackers in Long-Term Visual Tracking13
This paper provides a framework, named CoCoLoT, that combines the characteristics of complementary visual trackers to achieve enhanced long-term tracking performance and competes favourably with the state-of-the-art on the most popular long- term visual tracking benchmarks.
- 13
- Adaptive neural tree exploiting expert nodes to classify high-dimensional data13
This paper introduces a novel Neural Tree exploiting Expert Nodes (NTEN) to classify high-dimensional data based on a decision tree structure, whose internal nodes are expert nodes performing multi-dimensional splitting.
- Deep Super-Resolution Network for Single Image Super-Resolution with Realistic Degradations13
A deep SISR network that works for blur kernels of different sizes, and different noise levels in an unified residualCNN-based denoiser network, which significantly improves a practical CNN-based super-resolver for real applications.
- Mobile Food Recognition with an Extreme Deep Tree13
This work proposes an approach to food recognition which combines the strengths of different state-of-the-art classifiers, namely Convolutional Neural Networks, Extreme Learning Machines and Neural Trees, and shows that the proposed architecture can achieve good results even with low computational power, as in the case of mobile devices.
- From person to group re-identification via unsupervised transfer of sparse features12
The single person re-identification knowledge is transferred by means of a sparse dictionary learning to group re-Identification by obtaining a sparsity-driven residual group representation that is exploited to perform group re -identification.
- 12
- LowFormer: Hardware Efficient Design for Convolutional Transformer Backbones11
This work analyzed common modules and architectural design choices for backbones not in terms of MACs, but rather in actual throughput and latency, as the combination of the latter two is a better representation of the efficiency of models in real applications.
- Resolution based Feature Distillation for Cross Resolution Person Re-Identification11
A Resolution based Feature Distillation (RFD) approach that learns resolution invariant features by filtering the resolution related features from the final feature vectors that are used to compute the distance matrix to overcome the problem of multiple resolutions.
- 11
- 11
- Adaptive video communication for an intelligent distributed system: Tuning sensors parameters for surveillance purposes11
A system architecture for a video surveillance system with distributed intelligence over multiple processing units and with distributed communication over multiple heterogeneous channels is proposed and a new real-time technique for changing the video transmission parameters according to the available bandwidth is presented.
- 11
- 10
- 10
- Deep Iterative Residual Convolutional Network for Single Image Super-Resolution10
This work proposes a deep Iterative Super-Resolution Residual Convolutional Network (ISRResCNet) that exploits the powerful image regularization and large-scale optimization techniques by training the deep network in an iterative manner with a residual learning approach.
- 10
- 9
- An ADAS Design based on IoT V2X Communications to Improve Safety - Case Study and IoT Architecture Reference Model9
The paper presents the design process of the system architecture using IDEF modeling methodology and data flows investigations and proposes an approach for system architecture design of a safe intelligent driver assistant system using IoT communication.
- On filter banks of texture features for mobile food classification9
This work analyzes different texture filter banks to evaluate their performances and proposes a method to automatically aggregate the best features for food classification purposes, with particular emphasis on the computational burden of the system to match the limited capabilities of mobile devices.
- 9
- 9
- 9
- Pre-emptive Camera Activation for Video-Surveillance HCI8
A new information visualisation technique that aims to reduce the mental effort of security operators and allow them to follow the object of interest is introduced.
- 8
- 7
- 7
- Incorporating linear discriminant analysis in neural tree for multidimensional splitting7
The proposed hybrid classifier, neural tree with linear discriminant analysis called NTLD, adopts a tree structure containing either a simple perceptron or alinear discriminant at each node, which leads to a shallower tree together with better classification performance.
- SkelMamba: A State Space Model for Efficient Skeleton Action Recognition of Neurological Disorders6
This work introduces a novel state-space model (SSM)-based framework for skeleton-based human action recognition, with an anatomically-guided architecture that improves state-of-the-art performance in both clinical diagnostics and general action recognition tasks, and introduces a novel medical dataset for motion-based patient neurological disorder analysis to validate the method's potential in automated disease diagnosis.
- Consistent attentive dual branch network for person re-identification6
A novel Consistent Attention Dual Branch Network (CadNet) that has ability to model long-range dependencies (correlations) between channels as well as feature maps and significantly enhances the performance with respect to the baseline on the person re-identification benchmarks.
- 6
- 6
- A Deep Residual Star Generative Adversarial Network for multi-domain Image Super-Resolution5
A deep Super-Resolution Residual StarGAN (SR2*GAN), a novel and scalable approach that super-resolves the LR images for the multiple LR domains using only a single model.
- 5
- Person re‐identification by modelling principal component analysis coefficients of image dissimilarities5
An approach that exploits image dissimilarities is proposed, treating re-identification as a binary classification problem, and an approach on two publicly available benchmark datasets is evaluated and compared with state-of-the-art methods.
- Kernel‐based clustering5
This work proposes a novel clustering technique that exploits the geometric properties of normalised kernel spaces to automatically detect the correct number of clusters, thus avoiding the requirement of an initial estimate of this parameter, as required instead in many popular algorithms.
- 5
- 4
- 4
- 4
- A Distilled Model for Tracking and Tracker Fusion.4
This paper proposes a novel tracking methodology that takes advantage of other visual trackers, offline and online, and shows that the proposed algorithms compete with state-of-the-art trackers while running in real-time.
- An Exploration of the Interaction Between capsules with ResNetCaps models4
This paper starts from the CapsNet architecture to explore and analyse the interaction between the presence of features within certain, similar classes and shows that the exploitation of capsules interaction yields to performance improvements.
- 4
- 4
- How a visual surveillance system hypothesizes how you behave4
A novel method for hypothesizing the evolution of behavior is presented, defined by means of a novel clustering technique, which is tested in the context of an outdoor parking lot and allowed to speculate about any intention to act dangerously.
- 3
- 3
- 3
- Visible-Thermal Pedestrian Detection via Unsupervised Transfer Learning3
An unsupervised transfer learning framework is proposed for visible-thermal pedestrian detection tasks and demonstrates that the proposed method obtains state-of-the-art performance without any manual training labels on the target dataset.
- 3
- Self Attention based multi branch Network for Person Re-Identification3
A novel self attention based multi branch(classifier) network to directly model long-range dependencies in the learned features and encourages the learning to be independent of the feature map locations is proposed.
- Unsupervised Hashing with Neural Trees for Image Retrieval and Person Re-Identification3
This work proposes a simple yet effective unsupervised hashing framework, which exploits the powerful visual representation capabilities of deep architectures and combines this within a tree structure involving a multi-path scheme.
- Distributed person re-identification through network-wise rank fusion consensus3
A distributed person re-identification framework which brings in a weighted camera matching cost that measures the re-Identification performance between cameras in the network and a derivation of the distance vector algorithm that yields to network topology learning and allows us to prioritize and limit the cameras inquired for the re,identification.
- Selection of Temporal Features for Event Detection in Smart Security3
A formalized description of the features associated with each event and the linked strategy to define the event time-line are provided and preliminary tests carried out on videos related to a reference outdoor environment which validate the proposed strategy.
- Mobile Re-Identification Based on Local Features Analysis3
This paper presents a comparative study of facial local features for the task of person re-identification using embedded smart cameras in a realistic surveillance scenario and performs an empirical study of the performance of the existing schemes based on local facial features when fused with auxiliary information.
- Multi-sensor registration for objects motion detection3
A method to automatically register IR with visual image data is presented that uses geometric structures that are matched with a partial graph matching algorithm and an iterative method to map IR and visual sequences using the homography matrix between frames is introduced.
- Is Tracking Really More Challenging in First Person Egocentric Vision?2
A new benchmark study is introduced designed to disentangle challenges related to the first person perspective from those linked to the broader domain of human-object activity understanding, providing deeper insights into the true sources of difficulty in egocentric tracking and segmentation.
- 2
- 2
- Real Image Super-Resolution using GAN through modeling of LR and HR process2
A GAN-based SR approach with learnable adaptive sinusoidal nonlinearities incorporated in LR and SR models by directly learn degradation distributions and then synthesize paired LR/HR training data to train the generalized SR model to real image degradations.
- 2
- Surrogate Safety Measures Prediction at Multiple Timescales in V2P Conflicts Based on Gated Recurrent Unit2
A Gated Recurrent Unit-based system that aims to predict, up to 3 s ahead in time, the severity level of V2P encounters, depending on the current scene representation drawn from on-board radars’ data, and can be used to label pedestrians near accident events and to enhance existing intelligent driving systems.
- 2
- Video-Based Convolutional Attention for Person Re-Identification2
A Siamese framework is proposed in which video frames of the person to re-identify and of the candidate one are processed by two identical networks which produce a similarity score.
- An ensemble feature method for food classification2
This work adopts a Random Forest classifier to measure the performances of different texture filter banks and feature encoding techniques on three different food image datasets, and compares these classifiers with other feature-based state-of-the-art solutions.
- Application of balanced neural tree for classifying tentative matches in stereo vision2
A new application of the supervised learning based classifier in stereo matching using the speeded-up robust feature (SURF) matching is presented, and the chosen classifier is a balanced neural tree.
- Supervised Learning Based Stereo Matching Using Neural Tree2
A supervised learning based approach is presented to classify tentative matches as inliers or outliers obtained from a pair of stereo images using a balanced neural tree (BNT) to perform the classification task.
- 2
- 2
- Stereo Vision in a Network of Co-operative Cameras2
In this chapter, a smart framework is presented for the object localization on a given ground-plane test map using heterogeneous stereo vision and is useful in stereoscopic as well as in video surveillance applications.
- Event recognition with PTZ cameras2
A method for explicitly describing and matching complex events in terms of simpler elements is proposed, which proves the validity of the proposed event recognition technique.
- 2
- 2
- 2
- Better, But Not Sufficient: Testing Video ANNs Against Macaque IT Dynamics1
Current video models better capture appearance-bound dynamics rather than the appearance-invariant temporal computations expressed in IT, underscoring the need for new objectives that encode biological temporal statistics and invariances.
- Validation of Synthetic Megavoltage Computed Tomography (MVCT) for Dose Calculation in Radiotherapy Treatment Planning1
The results support the dosimetric feasibility of synthetic megavoltage images as a promising alternative to true megavoltage imaging for radiotherapy planning in patients with dental implants, pending validation in larger prospective cohorts.
- Beyond MACs: Hardware Efficient Architecture Design for Vision Backbones1
This paper experimentally demonstrates the shortcomings of MACs (Multiply Accumulate operations) as a predictor of execution time, and presents LowFormer, a novel vision backbone family that includes Lowtention, a lightweight alternative to Multi-Head Self-Attention.
- 1
- 1
- Lightweight Implicit Blur Kernel Estimation Network for Blind Image Super-Resolution1
This work designs a lightweight approach for blind super-resolution (Blind-SR) that estimates the blur kernel and restores the HR image based on a deep convolutional neural network (CNN) and a deep super- resolution residual Convolutional generative adversarial network.
- An Exploration of Target-Conditioned Segmentation Methods for Visual Object Trackers1
This paper proposes to extensively explore target-conditioned segmentation methods available in the computer vision community, in order to transform any bounding-box tracker into a segmentation tracker, and shows that such methods allow trackers to compete with recently proposed segmentation trackers, while performing quasi real-time.
- Generating Domain and Pose Variations between Pair of Cameras for Person Re-Identification1
A Domain and Pose Invariant Generative Adversarial Network (DPI-GAN) to generate images for both domain and pose variations capture to improve the performance of person re-identification.
- Unsupervised Smoke Detection in Normally Smoking Environments1
This work proposes an unsupervised solution that leverages on the concept of anomaly detection and uses a temporally smoothed representation to learn a robust "normality" model by means of a One-Class Support Vector Machine.
- Temporally Smoothed Anomaly Detection in Continuous Fluids1
This work uses different visual feature representation which are smoothed over time to compensate the temporal changes that might not be due to the presence of anomalies but to noisy measurements, etc. to learn a robust "normality" model by means of a One-Class Support Vector Machine.
- 1
- Distributed Mobile Computer Vision: Advances, Challenges and Applications1
This chapter provides an analysis of the recent advances of mobile computer vision, then a discussion of the current challenges that the community is currently dealing with, and introduces the next challenges in mobile vision where the mobile devices are part of a visual sensor network.
- 1
- 1
- Improved Statistical Techniques for Multi-part Face Detection and Recognition1
Experimental results on real images show the superiority of the proposed techniques with respect to the existing ones in both the detection and the recognition phase.
- Shadow Removal in Outdoor Video Sequences by Automatic Thresholding of Division Images1
A shadow detection algorithm able to remove shadows from the blobs of moving objects, using division images and Expectation-Maximization histogram analysis is proposed.
- 1
- An Autonomous Surveillance Vehicle for People Tracking1
In this paper, the problem of the surveillance and the security of indoor environments is addressed through the development of an autonomous surveillance vehicle (ASV) designed to perform the object detection by adopting an image alignment method followed by a change detection operation.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- From Wearable Sensor Networks to Markerless Motion Capture for Instrumental-Based Biomechanical Risk Assessment in Lifting Activities–
Compared with a wearable sensor network and an ML system in the evaluation of lifting tasks, ML consistently showed a closer agreement with reference benchmarks and a lower variability, and in terms of the automatic classification performance, ML–based kinematic variables yielded accuracy levels comparable to those obtained with the wearable system.
- –
- –
- UW-CVGAN: UnderWater Image Enhancement with Capsules Vectors Quantization–
This work introduces Underwater Capsules Vectors GAN UWCVGAN based on the discrete features quantization paradigm from VQGAN, and achieves feature quantization by exploiting the clusterization ability of capsule layer, making the model completely trainable and easier to manage.
- –
- –
- –
- –
- Biometric antispoofing on mobile devices–
A software-based spoof detection prototype for mobile devices, named MoBio_LivDet (Mobile Biometric Liveness Detection) that can be used in multiple biometric systems that allows user to choose “Security Level” (SL) against spoofing.
- Special Section Guest Editorial:Advances on Distributed Smart Cameras–
This PDF file contains the editorial “Special Section Guest Editorial:Advances on Distributed Smart Cameras” for JEI Vol.
- –
- Situation Graph Trees–
Situational Graph Trees (SGTs) represent such a modelling tool that can be used in complex environments to model behaviours by means of temporal and semantic relationships or specialization of fundamental concepts.
- Multiresolution Depth Map Estimation in PTZ Camera Network–
In this chapter, an active stereo vision system composed of two pan-tilt-zoom (PTZ) cameras is proposed for estimating multiresolution depth map for a large and complex scene.
- AVSS 2011 demo session: Smart Resource-Aware Multi-Sensor Network–
The invited talks are: How to Compare Alternative Architectures by Radia Perlman of Intel; Portals 4: Enabling Application/Architecture Co-Design for High-Performance Interconnects by Ron Brightwell of Sandia National Laboratories; and Electronic-Photonic Integration within Switches and Routers by Mike Watts of MIT.
- A hierarchical multi-sensor framework for event detection in wide environments–
Video surveillance applications often imply paying attention to a wide area, so different kinds of cameras are generally used, for example, fixed cameras, omni-directional cameras and pan, tilt and zoom cameras.
- –
- –
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 2026-10-14. 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.