What is a 'Collada' scene graph?

Question

What is a Collada scene graph?

Can I take a scene that I've downloaded from Google 3D warehouse and somehow convert it to a Collada scene graph?

Does the process somehow segment the scene?

Answer 1

COLLADA is the standard XML description language available for most if not all 3D applications. Google Earth is using COLLADA for the 3D models placed on to of the earth. Google Sketchup can import and export COLLADA models. Google Warehouse offer most if not all models in the COLLADA format. Google KMZ format is a ZIP file that contains COLLADA (.DAE) files(s), textures, and GIS placement information.

XML is a text based format, so you can look into a COLLADA document with any text editor, although a XML editor is highly recommended for large files, or for easier navigation. COLLADA is defined by its specification (in English and Japanese) and its Schema (.xsd) which enable COLLADA documents to be tested for validation by standard XML tools. Editors such as 'daenotepad' available on github will also provide help editing COLLADA documents based on the information stored in the schema.

A COLLADA document contains many different elements, one of the element is the < visual_scene > which contains the instance of geometries withing the the visual scene organized in a tree of locations. That is what is often referred incorrectly as the 'scene graph'. A scene graph is a run-time technology used by a lot of 3D rendering engines, which is similar to the but has many active run-time attributes such as 'switch' 'Level of detail' 'script' ... which COLLADA is not defining. COLLADA is merely defining the necessary data for enabling any applications, scene graph based or not, modeling tools or graphics engines to find the information they need a and communicate back and forth.