"Adaptive Simplify" for decimate modifier

I don't know what else to call this, but it would more or less be "Reverse adaptive subdivision." Where instead of subdividing a simplified mesh on render based on distance to camera, it un-subdivides a complex mesh based on distance to camera. The setting would probably fit right at home in the UI of the decimate modifier.

It would serve the same purpose as Adaptive Subdivision, it would just work in different use cases, such as when users are working with high poly sculpted models, but want to speed up GPU rendering.

I don't know if any of the code from Adaptive Subdiv would be able to be used, and it would depend on the developer's knowledge of the code whether this is worth looking into or not, but it was something that I have been thinking about recently.