Add ml_depth_pro example

[oxkitsune]

What

ml_depth_pro_example.mp4

This adds an external example for visualizing DepthPro using the new video logging api

Checklist

• I have read and agree to Contributor Guide and the Code of Conduct
• I've included a screenshot or gif (if applicable)
• The PR title and labels are set such as to maximize their usefulness for the next release's CHANGELOG

• PR Build Summary
• Recent benchmark results
• Wasm size tracking

To run all checks from main, comment on the PR with @rerun-bot full-check.

[Wumpf]

nice! Description lgtm, maybe a bit on the short side 🤷 Is there maybe a link to the paper that could be included as well?

I tried it with this old classic

For some reason it ended up only doing three frames hum.

[Wumpf]

looks like pixi run lint-rerun is unhappy. probably needs some NOLINT comments (there's now also BEGIN_NOLINT and END_NOLINT)

[oxkitsune]

Sure, I will add some more details to the description.

As for the error, it might have logged a text log for the error, I suspect it's HF limits.

I run a forward pass for each frame separately which might eat the ZeroGPU budget up too fast

[oxkitsune]

Getting this for each frame while running it:
😢

will look into batching it

[oxkitsune]

Lints pass, and batched inference works when I tested it locally on a 4090. I am out of ZeroGPU quota so I can't test it on HF for now