A5064727471- Face recognition and analysis
- Face and Expression Recognition
- Video Surveillance and Tracking Methods
- Sleep and Work-Related Fatigue
- Gaze Tracking and Assistive Technology
- Biometric Identification and Security
- Autonomous Vehicle Technology and Safety
- 3D Shape Modeling and Analysis
- Human Pose and Action Recognition
- Advanced Image and Video Retrieval Techniques
- Image Retrieval and Classification Techniques
- Hand Gesture Recognition Systems
- Blind Source Separation Techniques
- Advanced Vision and Imaging
- Medical Image Segmentation Techniques
- Anomaly Detection Techniques and Applications
- Advanced Neural Network Applications
- Emotion and Mood Recognition
- Computer Graphics and Visualization Techniques
- Speech and Audio Processing
- Reinforcement Learning in Robotics
- Robotics and Automated Systems
- Optical measurement and interference techniques
- E-commerce and Technology Innovations
- Advanced Adaptive Filtering Techniques
Sejong University
2016-2025
Polytechnic University of Timişoara
2019
Chungbuk National University
2013-2017
Korea Railroad Research Institute
2012
Smith-Kettlewell Eye Research Institute
2005
University of Cambridge
2002
Abstract Automatic anomaly detection is a crucial task in video surveillance system intensively used for public safety and others. The present adopts spatial branch temporal unified network that exploits both information effectively. has residual autoencoder architecture, consisting of deep convolutional neural network-based encoder multi-stage channel attention-based decoder, trained an unsupervised manner. shift method exploiting the feature, whereas contextual dependency extracted by...
This paper presents a study in which driver's gaze zone is categorized using new deep learning techniques. Since the sequence of zones driver reflects precisely what and how he behaves, it allows us infer his drowsiness, focusing or distraction by analyzing images coming from camera. A Haar feature based face detector combined with correlation filter MOSS tracker for detection task to handle tough visual environment car. Driving database big-data was constructed recording setup within...
Since fossil fuels are slowly depleting, bio and renewable energies now given more attention. The main purpose of this research is to investigate optimize the influencing parameters bioenergy production through transesterification process. application artificial intelligence (AI) in studies has become increasingly popular due its capability interpreting nonlinear relationships between inputs outputs for complex systems. Here, after conducting library carefully reviewing existing methods,...
Anomaly detection is to identify abnormal events against normal ones within surveillance videos mainly collected in ground-based settings. Recently, the demand for processing drone-collected data rapidly growing with expanding range of drone applications. However, as most aerial by flying drones contain dynamic backgrounds and others, it necessary deal their spatio-temporal features detecting anomalies. This study presents a transformer-based video anomaly method whereby we investigate...
Drones with obstacle avoidance capabilities have attracted much attention from researchers recently. They typically adopt either supervised learning or reinforcement (RL) for training their networks. The drawback of is that labeling the massive dataset laborious and time-consuming, whereas RL aims to overcome such a problem by letting an agent learn data its environment. present study utilize diverse within two categories: (1) discrete action space (2) continuous space. former has advantage...
This paper reviews the second NTIRE challenge on image dehazing (restoration of rich details in hazy image) with focus proposed solutions and results. The training data consists from 55 images (with dense haze generated an indoor or outdoor environment) their corresponding ground truth (haze-free) same scene. has been produced using a professional haze/fog generator that imitates real conditions scenes. evaluation comparison dehazed images. process was learnable through provided pairs...
This paper proposes a system that uses gaze direction tracking and head pose estimation to detect drowsiness of driver. Head is estimated by calculating optic flow the facial features, which are acquired with corner detection algorithm. Analysis driver's behavior leads three moving components: nodding, shaking, tilting. To track driver, we trace center point pupil using CDF analysis estimate frequency eye-movement.
This study presents a new method to track driver’s facial states, such as head pose and eye-blinking in the real-time basis. Since driver natural driving condition moves his diverse ways face is often occluded by hand or wheel, it should be great challenge for standard models. Among many, Active Appearance Model (AAM), Shape (ASM) are two favored We have extended Discriminative Bayesian ASM incorporating extreme cases, called Pose Extended—Active model (PE-ASM). Two databases (DB) used...
Training language models from scratch presents a critical challenge in Natural Language Processing (NLP), primarily due to the computational demands of pre-trained Large Models, which are predominantly trained on English corpora using extensive resources. While offering viable solutions, existing alternatives still rely heavily high-performance hardware. This work introduces different approach reducing algorithmic complexity Transformer-based architectures through U-Net Encapsulated...
Abstract We present a novel framework for efficient and robust facial feature representation based upon Local Binary Pattern (LBP), called Weighted Statistical Pattern, wherein the descriptors utilize straight-line topology along with different directions. The input image is initially divided into mean variance moments. A new moment, which contains distinctive features, prepared by extracting root k -th. Then, when Sign Magnitude components four directions using moment are constructed,...
Good performance in the sport of baseball shows that humans can determine trajectory a moving object and act on it under constraint rule. We report here neuronal activity supplementary eye field (SEF) monkeys performing an movement task inspired by baseball. In "ocular baseball," pursuit to target is executed or withheld based target's trajectory. found subset neurons SEF interpreted according Other specified at later time command pursue with eyes. The results suggest interpret sensory...
Discriminating between genuine and fake emotion is a new challenge because it in contrast to the typical facial expression recognition that aims classify emotional state of given stimulus. Fake detection could be useful telling how good an actor movie or judging suspect tells truth not. To tackle this issue, we propose model by combining mirror neuron modeling deep recurrent networks, called long-short term memory (LSTM) with parametric bias (PB), which features are extracted...
Although Radio Control (RC) has been a dominant device for controlling drone, it is known that fair amount of training period required to master it. One way sidestep such an RC-based control scheme would be utilizing either Kinect or Leap Motion sensor by which the user interacts with drone more naturally. In cases, however, pilot hang around since operating distance sensors rather short. this study, we propose new wearable human-drone interface embedded on Raspberry Zero, even novice can...