For 2.0
I had some thought about making a AI plugin to manage what a person would do. This quickly led me to think that to make this useful, assets, looks and environment assets would need to be tagged properly.
For example, a chair would need a "chair" tag, which leads to "an object that a person can sit on or do something with" -> "poses or animations you can do with a chair". Then seperate big area assets from objects into different categories. The area could have other tags like: "cold", "outside", "home", etc.
The best thing would be a way to share those tags with the community, so that if someone spends hours tagging assets, these can be updated with other users.
Then for scene browsing, something to maybe pack a scene bundle to take only one space, with a dropdown or extra menu to pick the scene you want.
Filters based on content to seperate for examples, scenes that contain only a person atom or no animation, to highlight "real scenes" where there is action. Also, the scene could automatically inherit tags from its content, which would make it really easy to find a scene based on what is inside without the scene author having to fill tags for the dependencies.
I had some thought about making a AI plugin to manage what a person would do. This quickly led me to think that to make this useful, assets, looks and environment assets would need to be tagged properly.
For example, a chair would need a "chair" tag, which leads to "an object that a person can sit on or do something with" -> "poses or animations you can do with a chair". Then seperate big area assets from objects into different categories. The area could have other tags like: "cold", "outside", "home", etc.
The best thing would be a way to share those tags with the community, so that if someone spends hours tagging assets, these can be updated with other users.
Then for scene browsing, something to maybe pack a scene bundle to take only one space, with a dropdown or extra menu to pick the scene you want.
Filters based on content to seperate for examples, scenes that contain only a person atom or no animation, to highlight "real scenes" where there is action. Also, the scene could automatically inherit tags from its content, which would make it really easy to find a scene based on what is inside without the scene author having to fill tags for the dependencies.
