skilld
All skills
nvidia avatar

/physical-ai-video-data-augmentation

@0482ebc
by NVIDIA Corporationnvidia/skills3.5k stars
424

Use when running video data augmentation and auto-labeling workflows on OSMO: flow selection, preflight, submit-time interpolation, monitoring, and output retrieval. Trigger keywords: video data augmentation, data enrichment, auto labeling, VDA demo, OSMO workflow, pseudo labeling.

Ask your Agent

Use this Skill: https://skilld.dev/gh/nvidia/skills/physical-ai-video-data-augmentation

This session only. Nothing lands on disk.

referencesflowse2e.md

≈647 tokens on demand. Your agent reads this file only when SKILL.md points to it.

E2E (Parallel)

Table of Contents

Runs full VDA graph with parallel original auto-labeling and augmentation after setup, then labels augmented outputs.

When to use

  • User requests full pipeline and prefers throughput.
  • Original labels and augmented labels are both needed in one run.
  • SR-gated sequencing is not required.

Graph

setup_group
  setup
      ▼
auto_labeling_original_group         augmentation_group
  pl_original_worker_0         ||      cosmos_worker_0
      ▼                                ▼
             auto_labeling_augmented_group
                   pl_augmented_worker_0

Inputs

Input Source Required by
Source video <storage_url>/datasets/<dataset>/<video>.mp4 setup + original AL + augmentation
Cosmos cache <storage_url>/data/models/cosmos_transfer cosmos_worker_0
Auto-labeling cache <storage_url>/data/models/auto_labeling pl_original_worker_0, pl_augmented_worker_0
VLM endpoint default in-cluster NIM or explicit override all workers
LLM endpoint default in-cluster NIM or explicit override all workers

Submit

Use the shared submit command and the common optional-overrides block in the SKILL.md "Submit (all flows)" section, with workflow YAML assets/configs/osmo/e2e.yaml. This flow runs original labeling and augmentation in parallel before augmented labeling, so the full canonical single-flag submit shape applies (including both cache URL values).

Output layout

<storage_url>/datasets/<dataset>-outputs/<run_id>/
├─ setup_b0/
└─ outputs/
   ├─ pseudo_labeled/<video>/
   ├─ augmented/<video>_aug0/
   └─ pseudo_labeled_augmented/<video>_aug0/

Troubleshooting

  • Completion evidence requirement: include a side-by-side input vs augmented video artifact, augmentation summary (setup_b0/configs/manifest.yaml sampled_vars for <video>_aug0), and auto-labeling artifact summaries for both outputs/pseudo_labeled/<video> and outputs/pseudo_labeled_augmented/<video>_aug0 using the workspace-local run copy under media/vda/runs/<run_id>/.
  • If original and augmentation groups do not run in parallel, verify no accidental task dependency was introduced between pl_original_worker_0 and cosmos_worker_0.
  • If augmented-label step cannot start, inspect cosmos_worker_0 output path and readiness first.

Source: SKILL.md on GitHub

Third-party checks

2 warnings3mo3 checks · Risk SAFE
  • Gen Agent Trust Hub3mo

    The skill is a workflow orchestrator for video data augmentation and auto-labeling on the NVIDIA OSMO platform. It manages the end-to-end pipeline, including credential verification, configuration generation, worker execution, and result retrieval. No security issues or malicious patterns were detected; all external dependencies and network operations are associated with trusted vendors and the skill's primary functionality.

  • Socket3mo

    3 alerts: gptAnomaly

  • Snyk3mo

    Risk: MEDIUM · 1 issue

Provenance

Signed by skilld at 0482ebc. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 4 months ago

Capability

Other metadata
metadata
{
  "owner": "NVIDIA",
  "service": "data",
  "version": "1.0.0",
  "reviewed": "2026-05-26",
  "author": "NVIDIA",
  "tags": [
    "physical-ai",
    "video-data-augmentation",
    "auto-labeling",
    "cosmos"
  ]
}

README badge

README badge for nvidia/skills/physical-ai-video-data-augmentation