Evaluation

Reflections of our progress, work and final deliverable

Overall

  1. Developed an AI-enhanced eye tracking solution using ordinary webcams, making it more accessible and cost-effective compared to existing solutions that rely on specialized external cameras.
  2. Achieved precise eye gaze tracking by implementing advanced techniques such as heat mapping, projective transformation, and calibration processes. This ensures accurate mapping of eye gaze coordinates from the camera's perspective to the corresponding screen coordinates.
  3. Designed the software with a focus on usability and performance, prioritizing intuitive user interactions, responsive cursor feedback, reliability across various environments, and compatibility with leading eye control software like Grid3 and EyeMine.
  4. Successfully tested the software with users having different levels of head movement control, demonstrating its adaptability and ease of use. The positive feedback from users like Sophia and Paul highlights the software's user-friendliness and potential impact.
  5. Garnered significant interest and enthusiasm from key partners such as Richard Cloudesley School, Mojang (Minecraft), Intel, and HP. The school's eagerness to adopt the solution and Mojang's interest in featuring it on their website underscore the software's potential for widespread adoption and impact on accessibility.
  6. Collaborated effectively with partners and incorporated their feedback to refine the solution, ensuring it meets the needs of the target users and is market-ready.

In summary, this project has made substantial contributions by developing an accessible, user-friendly, and technologically advanced eye gaze tracking solution. The successful tests, positive user feedback, and strong interest from key partners demonstrate the project's potential to significantly enhance accessibility and make a meaningful impact in the lives of individuals with limited mobility.

Summary of Achievements

The EyeGaze project, developed by Team 25, has made significant strides in creating an accessible and cost-effective eye tracking solution. By leveraging AI and computer vision techniques, the team has successfully developed a software that enables precise eye gaze tracking using only an ordinary webcam. This achievement is particularly noteworthy as it eliminates the need for specialized and expensive hardware, making eye navigation technology more accessible to a wider audience. The software's architecture focuses on key functional requirements such as eye detection, gaze estimation, cursor control, and click triggering. Additionally, the team has prioritized non-functional aspects, ensuring the software is intuitive, responsive, reliable, and compatible with existing eye-control optimized software. Through rigorous testing and user feedback, the team has demonstrated the software's ability to provide smooth and accurate cursor control. Users with varying levels of head movement control, such as Safiya and Paul, were able to intuitively calibrate and interact with the software, showcasing its adaptability and user-friendliness.

Critical Project Evaluation

While the EyeGaze project has made significant progress, there are areas that warrant further evaluation and improvement. The current accuracy of the gaze tracking, although enhanced through techniques like cursor smoothing, image enhancement, and increased calibration points, may still have room for refinement. Conducting more extensive user studies with a diverse range of participants could provide valuable insights into the software's performance across different user groups and help identify areas for optimization. Another aspect to consider is the software's robustness and reliability in various lighting conditions and environments. Testing the software under different scenarios, such as low-light or glare-prone settings, could help identify potential limitations and guide future improvements. Compatibility with existing eye-control optimized software is a key strength of the project. However, further evaluation of the seamless integration and performance when used in conjunction with popular tools like EyeMine would be beneficial to ensure a smooth user experience.

Contributions
Work packages Aleeyah Michael Sarah
Client liaison 33% 33% 33%
Target User Liaison 33% 33% 33%
Requirement analysis 33% 33% 33%
Research 15% 42.5% 42.5%
UI Design 70% 15% 15%
Programming 10% 80% 10%
Implementing MFC 100% 0 0
Testing 33% 33% 33%
Development blog 80 20 0
Website Editing 20% 20% 60%
Video Editing 0 100% 0
Overall contribution 30% 40% 30%
Roles Programmer, UIDesigner,
Implementer, Client liaison
Programmer, Tester,
Clientliason
Reseracher,Tester,
Website editor, Liason



Percentage breakdown
Work packages Aleeyah Michael Sarah
Client liaison 33% 33% 33%
Target User Liaison 33% 33% 33%
Requirement analysis 33% 33% 33%
Research 15% 42.5% 42.5%
UI Design 70% 15% 15%
Programming 10% 80% 10%
Implementing MFC 100% 0 0
Testing 33% 33% 33%
Development blog 80 20 0
Website Editing 20% 20% 60%
Video Editing 0 100% 0
Overall contribution 30% 40% 30%
Roles Programmer, UIDesigner,
Implementer, Client liaison
Programmer, Tester,
Clientliason
Reseracher,Tester,
Website editor, Liason

Future Work

Based on the achievements and evaluation of the EyeGaze project, there are several avenues for future work and improvement: Advanced Gaze Estimation Techniques: Exploring and incorporating state-of-the-art gaze estimation algorithms, such as deep learning-based approaches, could potentially enhance the accuracy and robustness of the eye tracking system. Expanded Compatibility: While the software is already compatible with some well-known eye-control optimized software, expanding the range of supported tools and platforms would increase its versatility and reach. User Customization: Implementing user-specific calibration profiles and settings could allow for a more personalized and optimized experience, catering to individual user preferences and needs. Multi-Platform Support: Extending the software's compatibility to various operating systems and devices, such as mobile platforms, could broaden its accessibility and usability. User Feedback and Iterative Refinement: Continuously gathering user feedback and conducting long-term studies would provide valuable insights for iterative improvements and ensure the software remains user-centric and responsive to evolving needs. Integration with Assistive Technologies: Exploring the potential integration of EyeGaze with other assistive technologies, such as speech recognition or gesture control, could create a more comprehensive and inclusive solution for users with diverse needs. By addressing these areas of future work, the EyeGaze project can continue to evolve and make significant contributions to the field of eye tracking and accessibility, ultimately empowering users with limited mobility to interact with technology more effectively and independently.