Real-Time Precise Human-Computer Interaction System Based on Gaze Estimation and Tracking

Noncontact human-computer interaction has an important value in wireless sensor networks. This work is aimed at achieving accurate interaction on a computer based on auto eye control, using a cheap webcam as the video source. A real-time accurate human-computer interaction system based on eye state recognition, rough gaze estimation, and tracking is proposed. Firstly, binary classification of the eye states (opening or closed) is carried on using the SVM classification algorithm with HOG features of the input eye image. Second, rough appearance-based gaze estimation is implemented based on a simple CNN model. And the head pose is estimated to judge whether the user is facing the screen or not. Based on these recognition results, noncontact mouse control and character input methods are designed and developed to replace the standard mouse and keyboard hardware. Accuracy and speed of the proposed interaction system are evaluated by four subjects. The experimental results show that users can use only a common monocular camera to achieve gaze estimation and tracking and to achieve most functions of real-time precise human-computer interaction on the basis of auto eye control.

A Systematic Assessment of YouTube Content on Femoroacetabular Impingement: An Updated Review

Background: Online video-sharing platforms such as YouTube have become popular sources of medical information for patients. However, concern exists regarding the quality of such non–peer reviewed content. In fact, a previous investigation found the majority of YouTube information related to femoroacetabular impingement (FAI) to be of poor quality. Purpose: To provide an updated assessment of the quality of FAI-related videos available on YouTube. Study Design: Cross-sectional study. Methods: The terms FAI, femoroacetabular impingement, and hip impingement were searched on YouTube, and exclusion criteria were applied to the first 100 results for each term. The diagnostic and treatment content of each video was graded and assigned a quality assessment rating based on a previously used rubric. Video characteristics (e.g. duration, views, “likes”) were compared using both quality assessment rating and video source. Results: A total of 142 videos were included in the final analysis. The most common video source was educational (48.6%), followed by physician-sponsored (30.3%). The majority of videos were graded as “somewhat useful” for both diagnostic and treatment content (59.4% and 61.6%, respectively); however, treatment content was rated “not useful” more often than diagnostic information (20.3% vs. 8.7%, respectively). Videos rated as “somewhat useful” received the most views per day on average, while educational videos were the most viewed by source (views and views per day). Educational videos had more views and likes on average than physician-sponsored videos ( P < .05), but all other comparisons of video characteristics by source were not significant. Video duration was the only characteristic found to vary significantly by quality assessment rating ( P < .001 for both diagnostic and treatment analyses), with higher-quality videos tending to be longer. Videos rated as “excellent” and “very useful” had mean durations >30 minutes but were viewed the least. Conclusion: The overall quality of FAI-related content on YouTube remains low. Clinicians should be familiar with medical information available to patients on the internet, as it can influence patients’ perspectives and shared decision-making processes. This review substantiates the need for more publicly available, high-quality video content regarding the diagnosis and treatment of FAI.

DRS ASSISTING UMPIRE USING COMPUTER VISION

A fair decision is crucial in any of the game to give justice to the game. Any wrong decision due to human misperception may fate the result of the game. Computer vision and Image processing techniques have been mentioned in the literature review which used multiple cameras for demonstration. This paper focuses on a system which helps in making the decisions to assist the umpire in taking the decisions such as no-ball, LBW i.e. Leg before wicket, Run out, stump out, etc with the help of smartphone camera of good quality. The Decision review system (DRS) aims to give decisions like run-out and stump-out. Tkinter is used to develop the GUI of DRS. Object classification and object recognition is implemented using Histogram of Gradients (HOG) and Support Vector Machine (SVM). To detect the cricket ball from the video we optimized and used frame subtraction, contour detection and minimum enclosing circle algorithms using OpenCV library. Linear regression and quadratic regression are used to track and predict the motion of the ball from video source. VPython is used for the visual representation.

Video source identification from MP4 data based on field values in atom/box attributes

Fast track article for IS&T International Symposium on Electronic Imaging 2021: Media Watermarking, Security, and Forensics 2021 proceedings.

Directive speech act in President Joko Widodo's speech related to handling coronavirus (Covid-19) in Indonesia (Pragmatic review)

Knowing the type of speech act is very important to understand to see the kind of speech act and its meaning and purpose behind the utterance. The purpose of this study is to describe the types, forms, and meanings or intentions of speech acts in the president's speech related to the handling of the coronavirus, especially with pragmatic studies. The type of research used is qualitative research. This study's data source is the video source contained in the upload of an account on Youtube. The data in this study are all speech acts, which are directive speech acts. The data collection techniques used are recording, observation, observation, and note-taking. The results showed: first, the directive speech acts consist of speech acts by asking, inviting, pleading, ordering, and prohibiting. Second, the speech act is in the form of an oral form, an utterance uttered by the president in his speech regarding the coronavirus's handling, which tends to be qualitative. Third, the purpose of this directive speech act has several meanings, namely to influence, invite, something, tell or explain something, expect attention, forbid, praise, criticize, and as a form of channeling complaints.

A Cross-Sectional Study of the Portrayal of Vocal Health in YouTube Videos

Purpose The current study examines metadata, upload source, type of informational content, understandability, and actionability of YouTube videos related to vocal health. Method A search for videos related to vocal health in YouTube was completed. Metadata (i.e., number of views, video length, thumbs-up, thumbs-down), upload source (i.e., consumer, professional, media), and type of informational content were identified. The Patient Education Material Assessment Tool for Audiovisual Materials was used to assess understandability and actionability of the 100 most frequently viewed videos. Results A total of 166 videos were obtained. The collective number of views was 19,799,299. A significant difference between video source groups was found for number of views and thumbs-down, but not for video length or thumbs-up. The YouTube videos were primarily educational, and a majority of the content focused on tips and techniques for professional voice users. Videos had adequate understandability (i.e., 71.5%) and actionability scores (i.e., 74%). Videos uploaded by consumers were superior to professional sources in actionability, but no difference was noted between video source for understandability. Conclusions Study insights about metadata, source, type of informational content, understandability, and actionability of YouTube videos may help professionals understand the nature of online content related to vocal health. Study implications and recommendations for further research are discussed.

Prevent, Detect, Respond, Mitigate Liquid Sodium Leakage, and Fire Accidents using AI

Liquid /Fluid Sodium is very risky in nature[1]. It gets ignited thusly when introduced to air or water. They recall repressions for smothering administrators. Social wellbeing and affirmation are basic. In many impelled organizations and present-day nuclear force stations, liquid sodium utilized as a basic coolant. This technique offers challenges to nuclear powerhouses. The fire remains an overall supporter of intensity plants. Research on capable fire ID and cautioning structures has become a fervently discussed issue in ventures. In many force stations and manufacturing plants, PLC or microcontroller-based controllers are utilized. Most of the customary frameworks will be rule-based. Now and again, they give counterfeit cautions. There is a need to naturally recognize outpouring and fire utilizing most recent advances like man-made reasoning. These days, most of the enterprises are fixed with CCTV or other observing frameworks. Our point is to utilize Video and Traditional way to deal with distinguish and moderate fire. More secure force stations created utilizing authentic mishap information and encounters. This paper discusses liquid metal spillage disclosure frameworks and abbreviates its state of-data achievement. Different fire revelation techniques explored, and their characteristics and inadequacies are highlighted. An AI based methodology is suggested that will manage the fire recognizable proof. Proposed system consolidates new advances like Computer vision, CNN calculation with the customary sensor or micro-controller or PLC based application to identify the smoke and fire segments. Sensors recognize the spillages, spillages of the fluid metals. Computer Vision and CNN calculations recognizes the nearness of fire from a current video source. This will limit the opportunity of spillages and different dangers.