Visual Imagination Model:
Input:
Input comes in as text.
For example, "A big bird above a tree."
Project Parse: Parse input words and determine imagination needs.
Visuo already does a bit of this.
Project Plan imagining: What goes in the imagagining and where does it go?
Visuo already does a part of this
The Oracle of Objects already does a part of this
Project conditional probabilities: Use co-occurrence of entities in DB to predict what and where.
Project Web Retrieve: Get images or 3d models from respositories.
Project python web retrieval:
make a simple python program that can make a search query on a web database (DB) and return the results.
Project place: Put the retrieved images, image pieces, or 3d Models in the appropriate places in the imagining.
Output:
1. a structured description of a fleshed out scene in Covlan
2. a image for the user to view, or a 3D virtual environment
Visuo
Input:
Text
Output:
Structured descriptions of quantitative magnitudes involved with the input text
in Covlan