Turn your videos into labeled training variants.
What it does
Produces augmented and automatically labeled videos through OSMO workflows, then retrieves the results and comparison evidence.
When to use it
Use for video augmentation and auto-labeling workflows on OSMO, not for container-internal tuning alone.
How to use it
Give it a video augmentation or labeling request; it uses your videos or defaults to demo footage and stages the results in the workspace.
Uses
Good to know
- cannot be undone: /srv/sdg/data/vda_inputs; OSMO
- acts without asking
What you need · 7
Requires the OSMO CLI on PATH for workflow submission, monitoring, and data access.
Requires an ONLINE OSMO GPU pool and GPU toleration/selectors in POD_TEMPLATE.
Access · 6
This skill
/srv/sdg/data/vda_inputs
Deletes existing demo MP4 files (irreversible)
OSMO
Starts GPU video processing workflows
Creates or replaces OSMO credentials (irreversible)