According to a recent study, people suffering of certain brain tumors are unable to dream anymore. However, the average person has about 1,460 dreams a year. Most people over the age of 10 dream at least 4 to 6 times per night during a stage of sleep, although we tend to forget 95 to 99 percent of our dreams.

This inspired us to create a playground that allows each of us to borrow and customize immersive dreams adapted in virtual reality.

Each user will be able to explore and customize its own DreamNA from a database of adapted dreams in VR.

According to a study by Amerisleep, the most common things American dream about are: falling (53%), being chased (50%), being back in school (37%), being unprepared for a test or for an important event (34%), flying (32%). The survey was done on over 2,000 people scattered throughout the U.S. Some researchers have found a correlation between the job of each person and the dream content.

Of course, the questions are endless as this is still a highly under-researched topic. Why do we dream? What is the meaning of our dreams? Do we have recurring dreams? With this project, we address questions rather than answers and to design a playful manner of accessing VR dream adaptations. On the long-term, we aim to create an arts-based research platform to explore collective dreaming.

What it does

DreamNA is a virtual playground that matches users with customized virtual reality dream adaptations.

We created an interactive website, that structures the user’s journey in three main chapters: Introduction: the viewer is introduced to the concept and the journey Questionnaire: the user answers five key-questions regarding his personality Dream matcher: Based on the answers of the user, the platform will provide a customized virtual reality dream. The VR video will be pulled out of a database of adapted dreams (based on advanced research statistics).

The VR dream may get different contents and different styles, depending on the answers of the user. For example, if the user feels sad, DreamNA will provide a VR dream that may cheer him up. If the user chooses a realistic dream, the platform will match him/her with a cinematic VR content.

For the moment we focussed on one user journey's prototype, aiming to perfect it as the users and database will grow.

How we built it

We used FFmpeg to decompose the cinematic VR video into frames, and then we used Cloudinary with Express to host and manipulate video and frames to create a unique dream experience. (ongoing development)

Challenges we ran into

It would have been nice to have the ability to manipulate video, instead of frames. The tools we used didn't support style transfer for videos, so we had to find a workaround by decomposing the video programmatically to apply the transform to each frame.

The visual effects we are interest in doing such as liquifying are not available. Maybe they can become available in a future release.

We spent time on shrinking the video in order to be able to load it into the tool, due to the file size.

Accomplishments that we're proud of

  • Succeeding to create a pioneering step towards a dream exploration as expressed through virtual reality.
  • Finding a way to customize cinematic VR contents with more ease.
  • The human and social layers of the project.
  • The potential it has to become a meaningful research and exploratory immersive playground on the long term.

What we learned

We have gained tech skills in Cloudinary, Express JS and FFmpeg. We have researched the topic extensively and made a long-term plan in order to reveal the complexity of it.

What's next for DreamNA

Short-term objectives for the current web platform:

  • add more visual effects to the VR dreams
  • add the possibility to customize sound
  • create an expanded database of adapted dreams in VR (based on extensive research) and user input (connecting the dreams with the feelings of the users as accurate as possible)
  • tech-wise, we aim to integrate EEG headbands, in order to allow each user to access and customize dreams according to their level of concentration and involvement.
  • content-wise, we aim to add an archive of room-scale fictional dreams in virtual reality, that would be much easier to customize. The content will be adapted after collected dream narratives.

On the long-term:

  • we aim to expand this concept intro a compelling transmedia journey and social experiment
  • the platform will integrate and interpret more data of dreamers (having geo-location features etc.)
  • creating the multi-player version of the project will allow us to merge customized collective dreams.

Built With

Share this project: