DA-3 training information

[nam1410]

Dear authors,
Thanks for your great work! Do you have any plans to release the training code or more detailed implementation information for Depth-Anything-3? I’m interested in reproducing your results. Thank you,

[anhquancao]

Hi! Not one of the authors, but in case it helps anyone here. We implemented the full training pipeline and losses for DA3 following the paper details. Hopefully is is a useful starting point:

https://github.com/valeoai/OccAny#occany-depth-anything-3--sam3-1

More context in #35 (comment)