[U22] - ReelSpace

Inspiration

The inspiration for ReelSpace emerged from our desire to push the boundaries of traditional movie-watching. As XR designers and developers, we envisioned a way to blend immersive technology with cinematic storytelling, creating an experience where users aren’t just watching a movie—they’re living it. We aimed to transform passive viewing into an active, engaging journey where the environment around the viewer adapts and enhances the story being told on screen.

What it does

ReelSpace revolutionizes the movie-watching experience by using generative AI to create adaptive, genre-specific environments. Users can choose between two modes:

1.  VR Mode: Offers full immersion in a 360-degree virtual environment, making users feel as if they are inside the movie.
2.  Window Mode: Provides an enhanced viewing experience where virtual elements seamlessly integrate with the user’s real-world surroundings, allowing them to watch the movie on their physical TV while enjoying an augmented environment.

The platform also supports real-time customization of these environments, enabling users to adjust lighting, effects, and even the arrangement of virtual objects, creating a personalized and deeply engaging cinematic experience.

How we built it

We built ReelSpace on the Oculus Quest 3, using the Unity Engine. The platform integrates OpenAI’s open-source tools to generate adaptive environments that match the movie’s genre. We also used the Meta SDK to ensure smooth performance and seamless integration with the Oculus hardware, making the experience both immersive and responsive to user input.

Challenges we ran into

Developing ReelSpace came with several significant challenges. One of the most complex was integrating generative AI to create adaptive environments that not only match the genre and mood of the film but also adapt in real-time based on user input. Ensuring these environments were both visually appealing and contextually appropriate required extensive fine-tuning of our AI algorithms.

Another critical challenge was balancing performance with graphics. Given the demanding nature of VR/AR environments, it was essential to maintain high visual fidelity without compromising performance, particularly on varying hardware capabilities across different VR headsets. This balancing act required optimizing our graphics engine and AI processes to ensure a smooth, immersive experience while maintaining the rich, detailed environments that enhance the user’s cinematic experience.

Accomplishments that we're proud of

We’re proud of how ReelSpace seamlessly blends AI with immersive technology to create a unique movie-watching experience. One of our major achievements is successfully integrating real-time user customization within AI-generated environments, allowing users to personalize their experience without breaking immersion. Additionally, we’re proud of overcoming the technical challenges of balancing high-fidelity graphics with the need for smooth performance across various VR platforms. These accomplishments position ReelSpace as a cutting-edge solution in the evolving landscape of home entertainment.

What we learned

Throughout the development of ReelSpace, we learned the importance of user-centric design in creating immersive experiences. Balancing innovation with usability is key—while advanced features are exciting, they must be accessible and intuitive for users. We also gained a deeper understanding of the complexities involved in real-time environment generation and the need for continuous optimization to maintain performance without compromising visual quality. Most importantly, we learned that collaboration across disciplines—design, engineering, and user experience—is essential to creating a product that is both technologically advanced and user-friendly.

What's next for ReelSpace

Looking ahead, we plan to expand ReelSpace by adding a broader selection of films, offering users more variety and ensuring that our adaptive environments can cater to different cinematic experiences. Our goal is to make the film-watching environment even more specific, with AI-generated models that adapt to the storyline as the movie progresses.

Additionally, we are exploring features that allow users to take screenshots of their favorite scenes. These screenshots would then be analyzed by the AI to generate corresponding virtual models, which users can use to decorate their virtual environments. This level of customization would give users even greater control over their movie-watching experience, making ReelSpace not just a platform for viewing movies, but a tool for creating personalized, immersive spaces that reflect their favorite moments in film.

Built With

• aipowered
• c#
• openai
• unity