POC implementation of Depth-anything to produce a 3D SBS video
This model costs approximately $0.13 to run on Replicate, or 7 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.
This model runs on Nvidia L40S GPU hardware. Predictions typically complete within 137 seconds. The predict time for this model varies significantly based on the inputs.
Proof of concept to use depth-anything on 2D videos to create depth frames to create a 3D SBS video
To convert the final 3D SBS video into a Spatial Video for the Vision Pro check out this Spatial CLI tool for apple silicon: https://blog.mikeswanson.com/spatial