Quilts are an image standard that Looking Glass uses to produce 3D experiences. This standard is used to describe both still and moving images (pictures and videos). Here is an example quilt picture:
Quilts serve a few purposes:
- to save and retrieve images displayed in a Looking Glass (similar to image & video screenshots for 2D monitors)
- as a compositing step in Looking Glass render pipelines found in our developer tools
- as a format to work against to manually produce lightfield photos / videos
Each tile in the quilt is a conventional 2D image of a scene. The bottom-left tile of the quilt is the leftmost view of the scene, and the top-right tile is the rightmost.
Standard formats for Looking Glass quilts are:
- 45 views: 9 rows and 5 columns
- 32 views: 8 rows and 4 columns
Non-standard formats are valid, but may not be as widely supported.
This standard may be applied to any conventional image or video filetype. The most common are
Looking Glass apps read from and write to metadata for files. The image standard is as follows:
- LKGNumViews: The number of views in the quilt.
- LKGRows: The number of rows.
- LKGColumns: The number of columns.
- LKGAspect: The aspect ratio of the quilt expressed as a fraction. This often encodes as a full floating point number and so is best to round to the second decimal point.
- LKGType: The type of file we're currently looking at. This can typically be inferred from the file format, but we have included "Image" and "Video" as a precaution.
These parameters are written to a JSON format to the UserComment section of exif data. When read properly, the data will take the following form: