Skip to content

Foundationpose pipeline vs Foundationpose setup

Foundationpose pipeline is an open-source agent skill for computer vision. Foundationpose setup is an open-source agent skill for computer vision.

  • Foundationpose pipeline is best for adapt the t-less dataset and run one scene through the tao depth engine.
  • Foundationpose setup is best for install the foundationpose perception pipeline on a gpu workstation.
What it is
Foundationpose pipeline is an open-source agent skill for computer vision.
Foundationpose setup is an open-source agent skill for computer vision.
Best for
Adapt the T-LESS dataset and run one scene through the TAO depth engine
Install the FoundationPose perception pipeline on a GPU workstation
Kind
Skill
Skill
Pricing
Open source
Open source
Platforms
Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf, Goose
Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf, Goose
GitHub stars
4K
4K
Last commit
Nov 2026
Nov 2026
License
Apache-2.0
Apache-2.0
Open source
Good for
Adapt the T-LESS dataset and run one scene through the TAO depth engine; Re-score a finished FoundationPose run at a new rerank cutoff while keeping the old report; Compare two pose summaries to see whether a 5 mm success rate changed at equal coverage
Install the FoundationPose perception pipeline on a GPU workstation; Fix a libcudart.so.13 loading error after moving the project folder; Build a TAO depth engine for adapted T-LESS scenes

What Foundationpose pipeline does

  • BOP dataset adaptation

    Converts supported BOP datasets into the rig layout the pipeline expects, with a registered adapter per dataset.

  • End-to-end pose inference

    Runs depth, segmentation and FoundationPose estimation, with or without pose ground truth for scoring.

  • Engine and input validation

    Smoke-checks the TAO depth engine for backend, normalization, fixed shape and valid-pixel fraction before a run.

  • Re-scoring completed runs

    Re-evaluates saved predictions with new rerank cutoffs, IoU thresholds or visibility bands without reloading models.

  • Run provenance checks

    Inspects engine, max-width and depth metadata so cached depth is only reused when it matches the current setup.

  • Batch evaluation sweeps

    Runs the pipeline across several datasets with optional continue-on-error and run status logging.

What Foundationpose setup does

  • Preflight host checks

    Verifies glibc, driver, CUDA toolkit, tooling and Docker GPU access before installation starts.

  • Pipeline installation

    Installs the perception pipeline, SAM3 and TAO Deploy into the product's Python venv.

  • TensorRT depth engine build

    Builds the FoundationStereo engine from ONNX using an adapted scene shape and precision choice.

  • Dependency conflict repair

    Resolves SAM3, TAO and numpy conflicts using inexact syncs and targeted installs.

  • Runtime path configuration

    Sets library paths and environment variables needed for CUDA and TensorRT libraries to load.

  • Staged verification

    Runs component verification scripts and a depth smoke test to confirm the setup works.

Foundationpose pipeline vs Foundationpose setup: questions

Is Foundationpose pipeline or Foundationpose setup cheaper?

Both are open source. Compare current plans on each official site.

What is the difference between Foundationpose pipeline and Foundationpose setup?

Foundationpose pipeline is an open-source agent skill for computer vision. Foundationpose setup is an open-source agent skill for computer vision.

Keep comparing