Sanjeev Kandel

Overview:Developed a computer vision system to recognize hand gestures from images, aimed at enhancing human-computer interaction (HCI) by enabling intuitive gesture-based commands.

Key Points:

• Problem Statement: Addressed the need for intuitive HCI methods beyond traditional interfaces, leveraging hand gestures as natural input modalities.
• Objectives: To develop a robust computer vision system capable of accurately recognizing and interpreting hand gestures for interaction with digital devices.
• Technologies Used: Python, OpenCV, TensorFlow, Keras, Scikit-learn.
• Data Collection and Preprocessing: Curated a dataset of hand gesture images encompassing a diverse range of gestures and lighting conditions, and preprocessed the data for training.
• Machine Learning Models: Employed convolutional neural networks (CNNs) and traditional machine learning algorithms to classify hand gestures from images.
• Model Training and Evaluation: Trained models using the collected dataset and evaluated performance through metrics such as accuracy, precision, recall, and F1-score.
• Application Development: Integrated the trained models into a prototype HCI system, enabling users to interact with digital interfaces through hand gestures.
• User Testing and Feedback: Conducted usability testing to assess the effectiveness and user experience of the gesture recognition system, iterating on the design based on user feedback.
• Results and Impact: Achieved accurate hand gesture recognition, demonstrating the potential for intuitive HCI methods and paving the way for future applications in diverse domains such as gaming, virtual reality, and assistive technology.
• Future Directions: Expanding the gesture vocabulary, improving real-time performance, exploring applications in immersive environments, and integrating the system with commercial devices for wider adoption.
• Team Collaboration: Collaborated with peers in AI, computer vision, and HCI domains, contributing expertise in machine learning and application development to the project's success.

• Problem Statement: Recognized the prevalent issue of mispronouncing names and its impact on effective communication and cultural sensitivity.
• Objectives: To create an application that offers phonetic spellings of names, helping users pronounce them accurately.
• Technologies Used: Python, TensorFlow, JavaScript, Flask, HTML/CSS, Docker.
• Data Collection and Preprocessing: Gathered diverse name data from various cultural backgrounds and preprocessed it for training models.
• Machine Learning Models: Implemented deep learning models utilizing natural language processing techniques to generate phonetic spellings.
• Model Training and Evaluation: Trained models using collected data and evaluated performance using metrics like accuracy and phonetic similarity.
• Application Development: Designed and developed the "Say My Name" application, incorporating a user-friendly interface and functionalities for easy name pronunciation.
• User Testing and Feedback: Conducted user testing to gather feedback on application usability and effectiveness, refining features based on user input.
• Results and Impact: Achieved improved pronunciation accuracy and received positive user feedback, enhancing cultural sensitivity and communication.
• Future Directions: Considering expanding language support, integrating additional features like audio pronunciation guides, and collaborating with cultural organizations for further enhancement.
• Team Collaboration: Collaborated with a multidisciplinary team, contributing to various aspects of the project from data collection to application development.

• Assisted in the design process for promotional materials which contributed to raising brand awareness among target audiences.
• Prepared comprehensive reports summarizing data analysis findings, informing key decision-makers of important trends and patterns.
• Managed social media accounts to boost company visibility, generating increased interest from prospective clients and customers.
• Improved project outcomes by conducting thorough market research that informed strategic marketing decisions.
• Contributed to a positive team environment by collaborating with fellow interns on group projects and presentations.