Browse
Recent Submissions
Item Flexible work arrangements and innovative work behavior(UMT,Lahore, 2025) Aqsa AmjadIn the modern fast changing workplace, innovation is very important for the success of organizations, especially in the dynamic IT industry. Although the past studies have recognized the positive impacts of Flexible Work Arrangements (FWAs), little concern has been shown on how FWAs affect the ability of employees to innovate. Based on the Self-Determination Theory (SDT), this paper explains the effects of FWAs on Innovative Work Behavior (IWB), through the mediating effects of Individual Dynamic Capabilities (IDC). The research design presents and empirically establishes a conceptual model with three major constructs, that are FWAs, IDC, and IWB. A structured questionnaire was used to collect data among 260 employees working at different positions in IT companies, located in Lahore, Pakistan. This is confirmed by the analysis, performed in SPSS, Process Macros by Hayes, and Sobel Test, that there is a positive and significant correlation between FWAs and IWB. Furthermore, findings indicate that IDC partially mediates this connection indicating that FWAs not only enable employees to be more creative by providing them with autonomy but also dynamic capabilities in them, hence contributing to their innovative work behavior. This study contributes to the existing body of literature as it bridges the gap between flexible work arrangements and innovative work behavior by introducing Individual Dynamic Capabilities as a mediator in the relationship. It also has practical implications since it demonstrates how flexible work systems can foster innovation by developing personal capabilities. It will be helpful for organizational leaders interested in increasing the innovation and agility of the workforce in the flexible work context.Item AI based body measurement app(UMT, Lahore, 2025) Aayet MirzaThis app takes users measurements as input, manually or through camera key point tracking and recording measurements through AI. It has an interactive user interface providing a tailor-in-hands experience. it lets a user choose through different designs, texture and prints , a totally customisable dress and providing Key features of the application include: • Real-Time Camera Feed: the app uses the device's camera to capture live video, which is processed frame-by-frame. • Gyroscope • Pose Estimation: the PoseNet model, integrated via TFLite, detects keypoints on the human body, including joints such as elbows, knees, and shoulders. • Overlay of Keypoints: Detected keypoints are visualized on the live camera feed, providing users with immediate feedback. • Measurement Calculation: Using the detected keypoints, the app calculates various body measurements, which can be used for fitness tracking, custom clothing fitting, and health monitoring. • Firebase Integration: the application uses Firebase for user authentication, real-time database storage, and cloud storage. User data, including measurements and preferences, is securely stored and synchronized across devices. • Dress Design Selection: Based on the calculated body measurements, users can select dress designs from a curated collection. the app provides recommendations for dress styles and sizes that best fit the user's measurements.Item VisionCart(UMT, Lahore, 2025) Muhammad Ahmed; Abdul Manan; Ahmed ManzoorItem Alif(UMT, Lahore, 2025) Abdullah Ahmad; Muhammad Muqeet; Sadia Jahangir; Abdullah FaisalALIF is a smart assistant robot, built by combining intelligent software with functional hardware. It is designed to make tasks easier, faster, and more accurate. More than just a machine, ALIF can handle multiple functions efficiently, allowing people to focus on creative and meaningful work. A big aim of ALIF is to help people and companies get things done quickly and without mess-ups. It's useful in tons of different areas, making it a great tool for all sorts of businesses. For instance, ALIF can assist doctors and nurses with paperwork, remind patients about their pills, or even just keep lonely people company. In factories, ALIF can take on boring or risky tasks that would wear people out. At home, it can help with chores, answer questions, and even keep you entertained. What's really cool about ALIF is how well it talks to humans. Unlike regular computers or robots that need special instructions, ALIF understands and answers in everyday language. You can just talk or type what you need in English (or another language it knows), and ALIF will chat back in the same way, making everything easy and smooth. Plus, ALIF can even pick up on how you're feeling by listening to your voice, looking atyour face, or reading your texts. This helps it respond in a way that fits your mood, making the whole experience more natural and personal. Another big plus is that ALIF can be changed to suit different people's needs. Whether you need a personal helper, a study buddy, a friendly customer service voice, or just someone to hang out with, ALIF can be programmed to do it. This makes it super useful in schools, offices, stores, and even homes. By putting together smart AI with physical parts, ALIF makes technology more interactive and easy to use. It's more than just a tool; it's a buddy that can help you out in both your work and personal life. As AI keeps getting better, ALIF could become a really important part of our daily lives, making work simpler, communication smoother, and technology more human.Item Pausify(UMT, Lahore, 2025) Minal Sheikh; Muhammad Sufyan Sarwar; Asad AliOverexposure to screens on smartphones and social media by children in the digital age affects not only mental health but also sleep, academic performance and exercise. The use of standard parental controls, like app blockers or screen time monitors, is frequently overlooked by children and can cause conflict between parents. We have a project in our final year called Pausify that is smarter and more subtle. Pausify passively reduces overall internet speed during specific times of the day, making it less appealing for children to spend extended periods on entertainment platforms like YouTube, Instagram, or Facebook. The slight discomfort gradually fosters improved habits without any direct conflict. Key features of Pausify include: • Time-based Internet throttling: Parents can schedule periods (e.g., 8–10 PM) where internet speed is slowed down on their child’s device. • Child Profiles: The schedule and settings of each child can be tailored to their daily routine. • Usage Analytics: By using usage analytics, parents can gain visual information about their usage patterns and the impact of throttling. Flutter is the foundation of Pausify, which incorporates Firebase for authentication, data storage and real-time syncing. The slowing system employs a VPN-based approach, which sends internet traffic through real time servers connected to the Internet via censored proxy networks. Pausify is designed with a user-friendly interface that is both intuitive and minimalistic. To sum up, Pausify advocates for a non-intrusive approach to digital balance. Through the use of psychological cues rather than restrictions, it can promote healthier screen habits for children and is an effective tool for modern parentingItem ReconSight(UMT, Lahore, 2025) Ahmad Bilal Bhatti; Anns Ijaz; Muhammad Mohsnain HaiderThis project focuses on the development of an advanced system for real-time object detection, tracking, and data retrieval, designed to enhance both efficiency and accuracy across various applications. By leveraging state-of-the-art deep learning frameworks such as YOLO and TensorFlow, along with powerful computer vision tools like OpenCV and DeepSORT, the system ensures precise object identification and seamless tracking in dynamic environments. To support scalability and real-time processing, the system integrates Flask and AWS, enabling optimized resource utilization and ensuring high-performance execution across different computing infrastructures. Additionally, it incorporates a robust academic timetable management feature, addressing scheduling complexities for students and administrators by offering automated scheduling, conflict resolution, and smart recommendations. Designed with a well-structured use case model, the system includes essential functionalities such as user authentication, image tracking, and efficient data search mechanisms. These elements contribute to a user-friendly and streamlined workflow, making the system highly adaptable to both academic and industrial applications. This work represents a significant step toward automating and enhancing real-time detection and tracking, offering practical, scalable, and intelligent solutions for a wide range of real-world applications.Item GATS(UMT, Lahore, 2025) Muhammad Sarosh Tahir; Muhammad Talal; Ammar Sarwar; Muhammad ShoaibIn the majority of schools, academic timetabling is a labor-intensive and time-consuming process. It is difficult for administrators to manage faculty availability, classroom assignment, and student scheduling, leading to inefficiencies and conflicts. Traditional methods are time-consuming and labor-intensive, which makes it difficult to react to unexpected changes such as faculty changes, room unavailability, or course changes. In order to overcome such limitations, we suggest GATS – Generative Ai Timetable Scheduler, an AI powered system with the ability to automate and optimize the generation of timetables. Driven by tools like Python, Pandas, and constraint-resolution libraries, GATS optimally manages resource allocation without sacrificing minimal scheduling conflict. Unlike static timetabling mechanisms, GATS is very flexible, with educational institutions having the capacity to adjust schedules dynamically according to shifting needs. With a welcoming interface and scalable design, GATS enhances study planning by reducing administrative burden, improving resource usage, and offering seamless scheduling experience. With intelligent automation, GATS makes timetable management an easy and streamlined process, enabling educational institutions to focus on delivering quality education rather than logistical problems.Item True Fit(UMT, Lahore, 2025) Saad Javed; Ali Adil WaseemThe artificial intelligence (AI) and complex algorithms of the mobile application True Fitdetermine when online apparel delivery consumers are likely to order intended-size ofclothing. Its platform ensures that the problem of online shopping is only an obstacle ofwhich size to choose from garments and this is because users are able to solve the problem.To this end, computer vision in conjunction with machine learning is used whenperforming the body point recognition in uploaded images as well as the application of themeasurement capture procedure. The site provides a solution to the common problem of online fashion buying as it connectsuser's related measurement data with standard size charts of popular fashion brands. Thebasic operations of True Fit like landmark identification and image processing occurs usingthe landmark identification tool utilizing MediaPipe, OpenCV platforms supported byTensorFlow machine learning libraries. The technological union allows for exactmeasurements on several types of hardware and same results. The technical correctness and the operating and fighting efficiency of the merchant and itscustomers are improved by True Fit. The program reduces costs and continues to produceproductive shopping processes due to it being able to achieve better length and widthmeasurements. Under such market conditions, the system empower online businesses tobuild their customer partner relations in order to provide individual users with with higherbusiness performance. The latest True Fit app uses artificial intelligence to bridge the gap between the onlineshopping experience and in stock fit experience in digital fashion as well as improve trustlevels and convenience and personalization features for consumers.Item Cheating detection system(UMT, Lahore, 2025) Ahmed Noor; Alishba; Wajeeha Arif; M TalhaExam cheating is a prevalent issue that undermines fairness in academia. Conventional approaches, where teachers or supervisors monitor students, are not guaranteed to be effective, particularly in large exam rooms. The purpose of this project is to create an AI- based system that assists in automatically detecting cheating. The system monitors students' posture motion while undertaking exams to detect abnormality. Deep learning models are employed by the system to detect four primary positions: normal, left, right, and back. By processing webcam images in real-time, the system is able to identify abnormal movements and notify examiners for investigation. For greater accuracy, we designed our own student body movement dataset rather than using online available data. We gathered, annotated, and processed the dataset very carefully to make it simulate real exam situations. A CNN-based system and image processing algorithms were employed for training to ensure robust movement detection. The system was tested and validated to monitor the system's accuracy and performance. The system is so designed that it can be operated with real-time exam monitoring software. It offers examiners an easy-to-use dashboard through which they can observe warnings if any student is caught attempting suspicious activities. This saves human effort, reduces errors, and provides a fair exam environment for all the students. In the future, we intend to enhance the system further by incorporating more data, employing eye tracking technology, and experimenting with other deep learning models to make it more accurate. We also intend to integrate it with sophisticated exam monitoring software so that greater institutions may utilize it. This AI-powered solution assists with ensuring fairness in exams, avoiding cheating, and making an exam environment more secureItem Course Craft (Ai tool)(UMT, Lahore, 2025) Hammad Ali, Eisham Azam and Syeda Mahroze BatoolCurriculum designing is a detailed and time-consuming process that must be done very carefully. Educators need to make sure course content is well aligned with defined learning objectives so that students achieve the required knowledge and skills. It is often a difficult task since it needs to organize lessons, connect the course topics and learning outcomes, and ensure compliance with educational standards. CourseCraft is an Artificial Intelligence (AI) tool aimed at simplifying and streamlining curriculum development. CourseCraft is based on the Outcome-Based Education (OBE) approach, where the focus is on obtaining measurable learning outcomes for students. CourseCraft employs cutting-edge Artificial Intelligence (AI) technologies like Named Entity Recognition (NER) and GPT models to read course content and sort it automatically into a formatted format. Among the most important features of CourseCraft is its functionality to generate course weekly outlines based on processing imported documents such as Excel and Word documents. It aligns the Course Learning Outcomes (CLOs) with Program Learning Outcomes (PLOs) so that every course significantly contributes to the overall educational program. It further assists in comparing student performance using these outcomes to enable teachers to determine areas that need improvement. The application uses machine learning and natural language processing (NLP) to make curriculum planning more accurate and efficient. Its simple interface allows teachers and educational institutions to personalize course structures as per their requirements. CourseCraft minimizes the time-consuming process of designing and modifying course content manually, freeing educators to teach more and perform less administrative tasks.Item AiRoomify Ai Interior Design Assistant(UMT, Lahore, 2025) Ali Hussain, Hammad Tariq, Nafay Mujtaba Ikhlaq and Syed Momin HasnainThe AI Interior Design Assistant is a web-based generative AI tool that allows users to easily redesign their homes. Through the upload of a photo of their room and inputting text explaining the changes they want to make, users can create AI-boosted room designs with Stable Diffusion for image processing and LLaMA 2 7B for smart text processing. The application is built with a Flask backend, React.js front end, and Tailwind CSS-styled for responsive user interface. The platform has also included e-commerce functionality, allowing users to purchase items such as sofas, beds, and tables within the app. The project aims at making interior design visualization and product choice easier and accessible through the potential of AI.Item Revolutionizing Shopping with AI Automation(UMT, Lahore, 2025) Muhammad Ahmed, Abdul Manan and Ahmed ManzoorRapid technological advancements have led to the world's evolution from villages to cities and now to smart cities, but retail shopping, particularly in grocery stores, still relies on traditional checkout methods, where customers must wait in long lines for extensive item scanning. This inefficiency causes customer dissatisfaction and emphasizes the need for an intelligent, automated solution to improve the shopping and checkout experience. By utilizing Artificial Intelligence of Things (AIoT) and integrating computer vision, and embedded system technologies, Vision Cart fills this gap and produces a smart shopping cart. Every cart has a raspberry pi 5 module, along with V1.2 camera module that uses cutting-edge computer vision techniques to collect real-time product photos and transmit them to a central server. These photos are processed by a Yolov8s model for product classification and recognition, guaranteeing real-time updates to a virtual cart that is shown on an attached screen. Furthermore, Vision Cart uses a Key Frame comparisons algorithm as tracking model, which works best in this situation, to keep bill updated of item additions and deletions.Item Skin Disease Detection(UMT, Lahore, 2025) Saliha Jamil, Tajallah Adil and Muhammad Arslan GondalTranslational research created the Skin Disease Detection project which developed a smart artificial intelligent system for automatic skin cancer and melanoma illness diagnosis. Advanced technology delivers better accessible dermatological diagnosis to users through a system that combines precision with efficiency. Through its user-friendly intelligent chatbot interface patients can efficiently submit information and upload photos related to their skin issues for smooth interaction. Through machine learning the submitted photos enable precise skin cancer diagnosis thus reducing misdiagnosis errors and facilitating rapid critical illness detection.Item Centralized AI Learning Management System(UMT, Lahore, 2025) Ch Ali Raza Mustafa, Muhammad Asad Ullah, Obaid ur Rehman, Muhammad AnsarItem Generative AI for Sketch transformation into Styled 2d images(UMT, Lahore, 2025) Dur-e-Sameen, Minahil Ijaz and Taifa MustafaThe pace of advancement in AI-based generative models has had a profound impact on the world of digital art, allowing machines to enhance creative processes and ambiguities in ways that often allows for a more efficient and imaginative way to create. The initiative that is introduced here takes advantage of these systems to bridge hand-drawn sketches to modern digital artwork through the development of an AI-powered platform that converts sketches to 2D images rich with style. Though text-to-image generating tools have proliferated in popularity, efficient sketch-to-image conversion tools remain relatively sparse, and even fewer allow for similar multiple artistic styles along with user's choice and control. To address this gap, we present here a learning-based system that combines Stable Diffusion with ControlNet for deep learning supports to learn the structure of the sketch and output information and quality in various artistic styles (i.e., realistic, cartooning, anime). To complete the task of the application, we created a front-end application using React and Vite and we developed a back-end API based on FastAPI for communicating with the model and handle task processing. We set up Supabase for user authentication, images stored, and user history.Item Automated Timetable Scheduler(UMT, Lahore, 2025) Awais Ali, Muhammad Mehdi Ishaq, Muhammad Kamran Hussain and Muhammad Qamer HassanTimetable scheduling in educational institutions is time-consuming and error-prone when done manually. The need to accommodate various constraints—room capacity, instructor availability, course prerequisites, and even non-teaching staff shifts—further increases complexity. Our research aimed to build an AI-based Automatic Schedule Maker capable of generating optimized, conflict-free timetables. In Phase 1, we began by experimenting with large language models (LLMs) via prompt engineering, to determine if they could autonomously generate timetables. While this approach showed promise, accuracy issues and limited adaptability led us to explore rule-based bots (like Edobot) and web-based solutions (like Unitime). Finally, we developed a Python-based scheduling algorithm that integrates AI-driven constraint-solving to automatically assign courses, rooms, labs, teachers, and staff shifts, all while preventing clashes. We validated this approach using Excel-based data input (courses, rooms, student counts), ensuring the solution is scalable to wider institutional needs. Our results show that the system substantially reduces manual workload, minimizes scheduling conflicts, and lays the groundwork for broader, AI-enhanced resource allocation in academic settings. Phase 2 expanded our research with a comprehensive dual-algorithmic framework, implementing both neural network-guided scheduling and genetic algorithm-based evolutionary optimization. Our genetic algorithm achieved optimal solutions with zero conflicts of test cases, demonstrating superior performance in complex constraint scenarios compared to traditional heuristic methodsItem LEGALMENTOR(UMT, Lahore, 2025) Faisal Javed, Daud Qaisar and Sara AzharItem InsightCart (AI- Based Smart Tracking System)(UMT, Lahore, 2025) Ghulam-Mohi-Ud-Din, Saim Khalid and Maheen SadafItem Nuskha AI(UMT, Lahore, 2025) Athar Naveed and Ibtisam AshrafWith the growing reliance on artificial intelligence in healthcare and daily management, Nuskha AI aims to provide a seamless and intelligent solution for medical prescription assistance and grocery management. This project integrates advanced AI-driven natural language processing (NLP) capabilities to help users understand medication details, dosages, side effects, and precautions while tracking their grocery needs. The system is designed to process text and image-based inputs, allowing users to upload prescriptions or grocery lists to efficiently extract and categorise relevant information. Additionally, Nuskha AI serves as a valuable tool for medical store merchants, enabling them to streamline inventory management, process order efficiently, and enhance customer service through AI-driven automation.Item FinAudit AI(UMT, Lahore, 2025) Sarim Zahid Saeed, Minhal Awais and Moiz AmjadIn pakistan, FinAudit AI is the first AI-powered audit platform that uses intelligent automation to rethink conventional financial auditing. To guarantee effectiveness, security and compliance in financial operations the platform integrates role-based access control (RBAC). AI driven audit workflows, OCR based documents classification and intelligent search driven by natural language procession (NLP) FinAudit AI automates key auditing procedures, such as document requirement production, audit report generation dynamic risk assessment, anomaly detection and findings creation through the integration of sophisticated AI agents. The platform gives minimum human error and manual labour by enabling proactive risk management and real-time audit monitoring.