A11yShape is changing the 3D modeling game for those with visual impairments. Liang He, assistant professor of computer science at the University of Texas at Dallas, created A11yShape. As a unique, innovative tool, this new resource opens thrilling creative and technical opportunities for blind programmers to expand their skills in modeling.
The tool features an advanced, user-friendly 3-D modeling web interface. It combines a code editor, an AI assistance panel, and a model panel into one easy-to-use interface. Together, these new tools grant users an unprecedented ability to wield the power of 3D design. It gives them timely, contextual feedback and aids their understanding of how iterative and complex code changes shape the resultant models.
A Multifaceted Approach to 3D Design
A11yShape is designed as three synchronized panels that play together to convey information. Code editor panel The code editor panel is where users will begin to write and edit code with ease. At the same time, the AI guidance panel provides live suggestions of their coding decisions. Additionally, the model panel shows the hierarchical organization and visual representation of the model you’ve built.
This addition is priceless for visually impaired programmers. It really empowers them to go through and explore on their own how different changes in code will impact their designs. When a user highlights a line of code or model component, A11yShape activates. It points out the matching features in each of the three side-by-side views. This aspect of the feature improves accessibility and helps encourage a more natural, hands-on way to learn about and engage with the design process.
“Provided [the blind and low-vision community] with a new perspective on 3D modeling, demonstrating that we can indeed create relatively simple structures.” – Participant in A11yShape testing
Throughout its development, A11yShape was piloted by four participants who possessed different levels of visual impairedness and programming background. Their feedback was instrumental in refining the user experience and ensuring that the platform met the needs of its target audience.
Quality Descriptions and Future Enhancements
Experience A11yShape produce beautiful resulting descriptions that have been highly rated. Scores for fidelity to prompt (geometric accuracy), clarity of generated response, and hallucination avoidance average from 4.1 to 5. This unprecedented quality on AI-generated descriptions is an important step for users seeking to easily convey intricate models to their audiences.
Looking forward, possible improvements for A11yShape involve the use of tactile displays and real-time 3D printing features. These new features will help us to connect the dots between the digital design world and real world reproduction. More importantly, they will enable users to interact with their models in novel and productive ways.
A11yShape could help them to implement AI-generated audio descriptions that are shorter and in turn, enhance the user experience. As technology continues to develop, these advancements will only continue to offer a wealth of potential for creativity and expression for visually impaired people.
“I want to design something useful and practical for the group.” – Liang He
Empowering Creativity Through Technology
While creating A11yShape, we were largely driven by a bigger purpose—making creative work possible for everyone using emerging technologies. As Stephanie Ludi points out, “People like being able to express themselves in creative ways… using technology such as 3D printing to make things for utility or entertainment.”
Liang He is quick to point out that A11yShape is bigger than just their specific project. It is a profoundly collaborative endeavor created to serve the richly diverse community. He states, “Not just something I created from my imagination and applied to the group.” This promise to inclusivity further emphasizes the need for developing tools that are available to everyone.