Hand Tracking Based Human-Computer Interaction Teaching System

In the modern era of computers, various new technologies have been arising. One such thing is a touchless application that is used or controlled aerially with hand gestures and movements. Augmented reality and virtual reality have come into use which is controlled by gesture controls. Applications that work with gesture controls have started targeting all kinds of users. Python libraries like MediaPipe and OpenCV are used in hand-tracking, palm detection and object detection. Our work aims in developing a virtual painter that helps young children to draw simple images and shapes of varying sizes. The tool recognizes the hand with hand and palm detector models of MediaPipe and capture the modes for selection and drawing using OpenCV library. In the covid pandemic where children are stuck at home and everything has become online, this tool helps them in practicing simple shapes virtually and also makes it interesting for them. The system is tested by drawing aerially with hands and using selection/drawing modes. It worked well with less time latency due to the inbuilt SSD algorithm used in MediaPipe. © 2022 IEEE.

Monitoring Engagement in Online Classes Through WiFi CSI

Due to the Covid-19 pandemic, people have been forced to move to online spaces to attend classes or meetings and so on. The effectiveness of online classes depends on the engagement level of students. A straightforward way to monitor the engagement is to observe students' facial expressions, eye gazes, head gesticulations, hand movements, and body movements through their video feed. However, video-based engagement detection has limitations, such as being influenced by video backgrounds, lighting conditions, camera angles, unwillingness to open the camera, etc. In this work, we propose a non-intrusive mechanism of estimating engagement level by monitoring the head gesticulations through channel state information (CSI) of WiFi signals. First, we conduct an anonymous survey to investigate whether the head gesticulation pattern is correlated with engagement. We then develop models to recognize head gesticulations through CSI. Later, we plan to correlate the head gesticulation pattern with the instructor's intent to estimate the students' engagement. © 2023 IEEE.

Wearable Technology based Face Alert System to Curb the Spread of Covid-19

Increasing people's perception of their habitual face-touching behaviour and ameliorating their acknowledgment of self-inoculation as a medium of transmission may assist to curb the spread of novel coronavirus (COVID-19). On average, human beings generally touch their faces 23 times per hour. Therefore, hand hygiene is an essential preventive measure to stop the spread of COVID-19. This motivates to introduce an alert mechanis m using wearable technology that aims to alert a person whenever he/she brings his/her hands close to the face. The proposed face alert system is based upon deep learning technique to forecast hand movements followed by face touching and imparts sensory response to alert end-user to stop the face touching activities. The proposed system employs IMU to get features belonging to different hand movements resulting in face touching. The data can be effectively classified using CNN where the filters help in extracting temporal features from IMU data. The prediction model based upon CNN is developed with training data from four thousand eight hundred trials recorded from forty participants. The trained dataset of hand movements activities is collected during day-to-day activities, e.g., walking, sitting, etc. Results demonstrated a forecast accuracy of 90% is obtained with 550ms of IMU data. In a research study, the psychophysical experiment is conducted to compare the response time for sensational observation methods, e.g., auditory, visual and vibrotactile. It has been observed that the response time is remarkably higher for visual (VF) and auditory feedback (AF) in comparison to vibrotactile feedback (VTF). Moreover, the rate of success is analytically lesser for visual feedback compared to vibrotactile and auditory feedback. Practically, results indicate a prediction of the movement of hand, and timely generation of sensational response in less than a second, so that one does not touch the face, and thus curbing of the spread of COVID-19. © 2022 IEEE.

Gesture - Controlled Contactless Switch With OLED Display

To avoid the chance of getting covid-19, it's vital not to touch surfaces as well as switches, door knobs and keys that are often employed by people. Hand movements in our world are the foremost well-liked non vocal ways in which of communication that are of agreeable significance. Gesture recognition is associated in Nursing interaction with human computers, normally used for functions of education, medicine and recreation. So, we came upon a contactless switch that works with hand gestures. Today with expanded mechanical progressions, switches also require refreshing with current technology. So, a non-contactless switch that works with sensors is the next step. Our keen contactless switch incorporates a sensor that is equipped for recognizing hand developments and interprets them into orders for controlling lights fans and different home machines. We are using Arduino IDE where we can create a setup function in which we can initialize the sensor and set the pin mode output or the light and fan control. © 2022 IEEE.

T2Snaker: a Robotic Coach for Table Tennis

The restrictions imposed by the Covid-19 pandemic has significantly affected all aspects of daily life, especially human contact. Accordingly, an essential aspect of human contact is for training and skill acquisition, which is difficult to conduct under such restrictions. Therefore, we developed T2Snaker, a table tennis training system that comprises a robotic appendage to guide user's hand movements within a VR environment. T2Snaker's novelty lies in its flexibility to guide users movements, yet as it is not directly attached to the user's limbs, it does not impose restrictions on their movements like traditional exoskeleton systems. We explain the implementation specifics of T2Snaker and discuss its preliminary evaluation that focused on table-tennis skill acquisition. The results show that T2Snaker has high potential in skill acquisition, and users praised is ability to guide their movements and proposed various potential application domains. We discuss some design insights based on our work and present future research directions. © 2022 Owner/Author.

Development of an IoT Based Domestic Monitoring and Alerting System for Paralysed Patients

We might come across lots of paralyzed patients in our day-to-day life. The cause of the condition could be due to trauma because of accidents, stroke, polio etc., Depending on the nerve response to the brain and spinal cord, the specific parts of the body or even the whole body become paralyzed accordingly. To make distance communication between the patient and caretaker easier, we have implemented an IoT based system to send and receive an SMS by using LabVIEW Software. The patient can send the SMS easily by hand movements. The message will be sent according to the direction of the patient's hand movement. This Software is used to send, read the data to the cloud and message will be sent to the caretaker's mobile simultaneously. We have also used GSM in our project to send the message as SMS. The main motto of this project is to send the message to the caretaker as fast as possible. Besides paralysed individuals, visually impaired persons and Covid patients in hospitals who are unable to use a cell phone and need to make contactless communication with physicians, can utilize it. This gadget can be used for a multitude of long-distance communications. The prototype is relatively inexpensive and portable. © 2021 IEEE.

Development of Real-time Handwashing Assessment using Video Analysis

The world communities have suffered from the COVID-19 pandemic for the last two years. Even though many countries have started to normalise the situation, the COVID-19 still becomes a severe threat in the future. Healthy habits, such as complete and frequent handwashing, still need to be practised. These habits can minimise the transmission risks. The paper proposed a single-board computer system that aims to assess the handwashing steps. The standardised handwashing procedure is used to validate the acquired video of hand movement. The system is installed in a Raspberry Pi and receives video data from the connected mini camera. The deep learning model is implemented to provide classification capabilities. The assessment result is summarised according to the movement completeness and total duration. The testing stages found that the proposed system can provide accuracy and F1-score values of 82.55% and 86.66%, respectively. © 2021 ACM.