New to Claude Skills? Learn how to install them →

nvidia on GitHub

Video Calibration

OfficialFree

Calibrate datasets from pre-recorded MP4 videos easily.

by nvidia2.8k stars on nvidia/skills
Updated Aug 7, 2026
Get this skill

Free · Opens the source repo

What Video Calibration does

The Video Calibration skill is designed to facilitate the calibration of datasets using pre-recorded MP4 video files through the AutoMagicCalib REST API. This skill is particularly useful for users who have local video files and want to streamline the calibration process without needing to engage with command-line interfaces or Docker setups. By simply activating the skill and providing the necessary video files and microservice URL, users can initiate the calibration process with minimal hassle.

To use this skill, users must have their MP4 files organized and named correctly, as the skill requires them to follow a specific naming convention (e.g., cam_00.mp4, cam_01.mp4). The skill automates the upload of these files to the AMC backend, ensuring they are processed in the order they are uploaded. Users can also provide optional parameters such as ground truth data and focal lengths to enhance the calibration results.

The skill operates by first creating a project on the AMC backend, followed by the upload of the video files, and finally, it resolves any necessary calibration settings and alignment data. If certain required files are missing, the skill provides a fallback to the user interface, guiding users through the necessary steps to complete the calibration process. This approach allows users to maintain control over their data while leveraging the power of the AMC microservice.

This skill is ideal for developers and researchers working with video datasets who require a straightforward method to calibrate their data without delving into complex setups or configurations. It is particularly suited for environments where users already have the AMC microservice running and need to process video files efficiently.

When to use it

Use this skill when you have pre-recorded MP4 video files that need calibration through the AMC microservice.

When not to use it

Do not use this skill for live RTSP streams; instead, use the amc-run-rtsp-calibration skill for those scenarios.

What you can build with it

Calibrating Research Videos

A researcher has multiple MP4 videos from an experiment and needs to calibrate them for analysis. This skill automates the calibration process via the AMC API.

Streamlining Video Processing Workflows

A developer wants to integrate video calibration into a larger application workflow. This skill allows for easy calibration of datasets without manual intervention.

Preparing Datasets for Machine Learning

A data scientist has collected video data for a machine learning project and needs to calibrate the footage. This skill provides a straightforward method to prepare the data.

How to install Video Calibration

View source

1. Install with the skills CLI

npx skills add nvidia/skills/amc-run-video-calibration --agent claude-code

2. Or install it manually

Download the skill folder and drop it into ~/.claude/skills/ for all projects, or .claude/skills/ to scope it to one repo. Restart Claude Code so it picks up the new skill.

Anthropic's agentic coding CLI, and the reference implementation of Agent Skills. Drop a skill folder into ~/.claude/skills and Claude Code loads it automatically whenever a task matches the skill's description. Claude Code docs

Inside SKILL.md

Written by nvidia

Skill: Calibrate from Video Files

When to Use This Skill

Activate this skill when the user has pre-recorded MP4 files and wants to calibrate them via the AMC REST API. Typical prompts:

  • "calibrate my videos" / "run AMC on these videos"
  • "calibrate from video files"

Drives calibration through the REST API on user-supplied pre-recorded MP4 files — no CLI scripts or Docker bind-mounts required, just a running microservice and your files.

Do not use this skill for live RTSP streams or rtsp://... URLs; route those requests to skills/amc-run-rtsp-calibration/SKILL.md.

Prerequisites

  • AMC microservice and UI running (follow skills/amc-setup-calibration-stack/SKILL.md)
  • You know the microservice URL (e.g. http://<HOST_IP>:<MS_PORT>) and UI URL
  • Video files locally as cam_00.mp4, cam_01.mp4, … time-synchronized, ~1920×1080
  • Python 3 with requests

Data Privacy

Video files uploaded via this skill are transmitted to the AutoMagicCalib backend (REST endpoint). Only use this skill when the backend is deployed on a trusted platform / network.

What to Ask the User

Required

(Video-file naming and the microservice URL are specified under Prerequisites above — collect the inputs below.)

  1. Videos directory — the folder the skill globs for cam_*.mp4, uploaded sorted alphabetically.
  2. Microservice URL
  3. Project name — short descriptive string

Auto-Detected (ask only if not found)

The script searches the videos dir, its first-level subdirectories, and its parent. If exactly one match is found, it is used; otherwise the script prints the searched locations and continues to explicit path or UI fallback:

FileCandidate filenamesUI fallback
Calibration settingssettings.json, config.json, calibration_config.jsonUI Step 3: Parameters
Alignment JSONalignment_data.jsonUI Step 4: Alignment
Layout PNGlayout.pngUI Step 4: Alignment

Posting the settings file replaces UI Step 3 and may pin the detector (resnet/transformer), which is passed to /calibrate separately — see Step 4.

Optional

  1. Ground truth zipGT.zip with _World_Cameras_Camera_XX/ folders (enables evaluation metrics)
  2. Focal lengths — one per camera, e.g. 1269.0, 1099.5, 1099.5
  3. Detector typeresnet (default, fast) or transformer (slower, better under occlusion)
  4. Run VGGT refinement? — if VGGT is ready after AMC completes, ask the user whether to run refinement (see setup skill)

See root README.md "Custom Dataset" section for input-video guidelines and ground-truth format.


Instructions

All endpoints below are implemented end-to-end in the Complete Python Script — the prose is the workflow plus the decisions the agent must make; the script is the authoritative runnable.

Step 1 — Create Project

POST /v1/create_project (form field project_name) → save the returned project_id.

Step 2 — Upload Videos (required)

POST /v1/upload_video_files/<project_id> (multipart files). Upload sorted alphabetically — the server assigns camera indices by upload order.

Step 3 — Resolve Local Files (Auto-Scan, Ask, or UI)

For each of calibration-settings, alignment, and layout, run this resolution:

  1. Auto-scan VIDEO_DIR, one level of subdirectories under VIDEO_DIR, and VIDEO_DIR.parent for the candidate filenames (table above).
  2. If exactly one match, use it and print what was found.
  3. If zero or multiple matches, print the searched locations, then ask the user for an explicit path using the host's question mechanism; if none is available, ask in chat and wait. If they don't have the file, mark it for UI fallback.
  4. UI fallback: tell the user to complete the corresponding UI step; wait for confirmation; for alignment/layout also verify files landed in projects/project_<id>/manual_adjustment/.

Step 4 — Upload Resolved Files

Upload each file resolved locally:

FileEndpointNotes
Calibration settingsPOST /v1/config/<project_id> (JSON, posted as-is)Replaces UI Step 3 (rectification, bundle-adjustment, evaluation, detector, …). Non-2xx is surfaced — never silently fall back. Skip on the UI-fallback path.
AlignmentPOST /v1/upload_alignment/<project_id> (alignment_data.json)
LayoutPOST /v1/upload_layout/<project_id> (layout.png)
Ground truth (optional)POST /v1/upload_gt_file/<project_id> (GT.zip)Enables evaluation metrics
Focal lengths (optional)POST /v1/upload_focal_length/<project_id> (repeated focal_length=)Overrides GeoCalib estimates

After a successful settings POST, parse the file for "detector" / "detector_type" — if it's "resnet" or "transformer", use that value for the /calibrate call in Step 7 (detector is a separate API parameter, not consumed by /config).

Step 5 — UI Fallback (only for files the user doesn't have locally)

If any of settings / alignment / layout was not resolved in Step 3, direct the user to the appropriate UI step:

  • Settings missing → "Open UI project <project_id>, go to Step 3: Parameters, tune via the settings dialog (or accept defaults), click Save." Also: before the /calibrate call, ask the user which detector to use (resnet or transformer) using the host's question mechanism; if none is available, ask in chat and wait. UI Step 3 does not cover detector choice.
  • Alignment or layout missing → "Open UI project <project_id>, go to Step 4: Alignment, upload layout, mark correspondence points, click Save."

Wait for user confirmation. For non-interactive script runs, provide the needed files up front; the script exits with a clear message rather than waiting on input. For alignment/layout, verify on disk before continuing:

: "${REPO_ROOT:?set REPO_ROOT to the auto-magic-calib checkout. Run amc-setup-calibration-stack Step 0b first.}"
grep -q "AutoMagicCalib" "$REPO_ROOT/README.md" 2>/dev/null && grep -q "auto-magic-calib-ms" "$REPO_ROOT/compose/ms/compose.yml" 2>/dev/null || { echo "ERROR: REPO_ROOT is not an auto-magic-calib checkout: $REPO_ROOT" >&2; exit 1; }
# Resolve PROJECT_DIR from the Compose environment file (default: projects/ at repo root).
COMPOSE_ENV_BASENAME="env"
COMPOSE_ENV_FILE="$REPO_ROOT/compose/.${COMPOSE_ENV_BASENAME}"
PROJECT_DIR_REL=$(grep ^PROJECT_DIR "$COMPOSE_ENV_FILE" 2>/dev/null | cut -d= -f2 | tr -d '[:space:]')
HOST_PROJECTS=$(cd "$REPO_ROOT/compose" && realpath "${PROJECT_DIR_REL:-../../projects}")

ls "$HOST_PROJECTS/project_<project_id>/manual_adjustment/"
# Expected: alignment_data.json, layout.png

Step 6 — Verify Project

POST /v1/verify_project/<project_id> → must return {"project_state": "READY"} before calibrating.

Step 7 — Start Calibration

Confirm the plan before calibrating. Whether the settings file and detector were auto-detected or asked, present a short summary and get explicit user confirmation before POST /calibrate using the host's question mechanism; if none is available, ask in chat and wait. The resolved values are the defaults, so confirming is one click, but the user can switch the detector or skip an auto-detected settings file. The standalone Python script prints the same plan and prompts only when stdin is interactive. Summarize:

  • Detectorresnet or transformer (the value to be sent).
  • Calibration settings — the file being applied (path), or "defaults" if none.
  • Optional overrides — ground-truth zip and focal lengths, if any.
POST /v1/calibrate/<project_id>
Content-Type: application/json

{"detector_type": "resnet"}

Step 8 — Poll for Completion

GET /v1/get_project_info/<project_id> every 10 s — project_info.project_state goes RUNNINGCOMPLETED (or ERROR, pull the log). Typical time: 10–60 min depending on video length and detector.

Step 9 — Get Results

GET /v1/result/<project_id>/evaluation_statistics (only if GT was uploaded; includes Average L2 distance(m) and Average reprojection error 0(px)), and GET /v1/amc/calibrate/<project_id>/log for the calibration log.

Status Fields from get_project_info

project_info.project_state is the AMC calibration lifecycle for the project: RUNNINGCOMPLETED (or ERROR).

project_info.vggt_state is also per-project, a project-scoped VGGT refinement lifecycle rather than a direct global service or model-load status. A newly created project can report vggt_state: "INIT" even when the VGGT model is present and mounted. The expected VGGT lifecycle is INITREADY after AMC calibration completes → RUNNING while VGGT refinement runs → COMPLETED (or ERROR).

Use vggt_state == "READY" only as the gate for optional VGGT refinement in Step 10. Interpret INIT on a new or uncalibrated project as normal project state. If AMC calibration is complete and the project remains in a non-ready VGGT state, confirm VGGT setup and model availability with the setup skill checks and service logs.

Step 10 — (Optional) VGGT Refinement

After AMC calibration completes, read vggt_state from GET /v1/get_project_info/<project_id>.

  • If the project reports vggt_state == "READY", ask the user whether to run VGGT refinement using the host's question mechanism; if none is available, ask in chat and wait.
  • If the user confirms, POST /v1/vggt/calibrate/<project_id>, poll vggt_state via get_project_info, then GET /v1/vggt_results/<project_id>/evaluation_statistics.
  • If VGGT is not ready, skip refinement and explain that the user can set up VGGT with amc-setup-calibration-stack and rerun this optional step later.

The standalone Python script prompts only when stdin is interactive. In non-interactive runs, set RUN_VGGT = True to opt in; otherwise the script prints that VGGT is ready and continues without blocking.


Complete Python Script

Use the bundled script from the amc-run-video-calibration skill package, not from the auto-magic-calib repo root. If the user points the agent at this skill folder directly instead of installing it, set AMC_VIDEO_SKILL_DIR to the directory containing this SKILL.md, or run the command from that directory. Set BASE_URL, PROJECT_NAME, and VIDEO_DIR; optional env vars are CONFIG_FILE, ALIGNMENT_JSON, LAYOUT_PNG, GT_ZIP, FOCAL_LENGTHS, DETECTOR_TYPE, RUN_VGGT, REPO_ROOT, and PROJECTS_DIR. The script implements UI fallback, plan confirmation, VGGT prompt/opt-in behavior, polling, and refined statistics retrieval.

# Optional but recommended: REPO_ROOT points to the auto-magic-calib checkout.
# PROJECTS_DIR can be set explicitly when project outputs live elsewhere.
if [ -z "${DEEPSTREAM_REPO_ROOT:-}" ] && [ -n "${REPO_ROOT:-}" ] && [ -d "$REPO_ROOT/../../skills/amc-run-video-calibration" ]; then
  DEEPSTREAM_REPO_ROOT="$(cd "$REPO_ROOT/../.." && pwd)"
fi

SCRIPT_PATH=""
for candidate in \
  "${AMC_VIDEO_SKILL_DIR:+$AMC_VIDEO_SKILL_DIR/scripts/run_video_calibration.py}" \
  "$PWD/scripts/run_video_calibration.py" \
  "${DEEPSTREAM_REPO_ROOT:+$DEEPSTREAM_REPO_ROOT/skills/amc-run-video-calibration/scripts/run_video_calibration.py}" \
  "$PWD/skills/amc-run-video-calibration/scripts/run_video_calibration.py" \
  "$HOME/.claude/skills/amc-run-video-calibration/scripts/run_video_calibration.py" \
  "$HOME/.codex/skills/amc-run-video-calibration/scripts/run_video_calibration.py" \
  "$HOME/.cursor/skills/amc-run-video-calibration/scripts/run_video_calibration.py"; do
  if [ -f "$candidate" ]; then
    SCRIPT_PATH="$candidate"
    break
  fi
done

[ -n "$SCRIPT_PATH" ] || {
  echo "ERROR: could not find amc-run-video-calibration/scripts/run_video_calibration.py" >&2
  echo "Set AMC_VIDEO_SKILL_DIR to the amc-run-video-calibration skill directory, or run this block from that directory." >&2
  exit 1
}

python3 "$SCRIPT_PATH"

Success Criteria

  • project_state == "COMPLETED" after polling.
  • If manual alignment was used: projects/project_<id>/manual_adjustment/ contains alignment_data.json + layout.png.
  • If GT was uploaded: evaluation returns typical thresholds:
    • Average L2 distance(m) < 1.5
    • Average reprojection error 0(px) < 5
  • No ERROR state.

Key Output Files (on server)

projects/project_<project_id>/
├── manual_adjustment/
│   ├── alignment_data.json
│   └── layout.png
├── output/
│   ├── single_view_results/cam_XX/
│   │   ├── camInfo_hyper_XX.yaml
│   │   └── trajDump_Stream_0_3d.txt
│   └── multi_view_results/BA_output/results_ba/
│       ├── initial/camInfo_XX.yaml
│       └── refined/camInfo_XX.yaml          # ← final calibration
└── calibration.log

Troubleshooting

IssueFix
verify_project state not READYConfirm videos uploaded and alignment + layout are present (either via API or via UI manual alignment)
Manual alignment files missing after UI stepUser didn't click Save; also verify projects/project_<id>/manual_adjustment/ exists
Calibration stuck RUNNING > 90 minGET /v1/amc/calibrate/<id>/log — usually insufficient tracklets (scene too static). See "Custom Dataset" guidelines in root README.
Immediate ERROR stateCheck video naming: must be cam_00.mp4, cam_01.mp4, … contiguous
Low L2 but high reprojectionProvide explicit focal_length override via Step 3
VGGT stays non-ready after AMC completesINIT is expected for a new project. After AMC calibration reaches COMPLETED, the project should transition to READY before optional VGGT refinement when VGGT is configured. If refinement is required and the state remains INIT or otherwise non-ready, confirm VGGT setup and model availability with setup skill Step 2 and MS logs.
Upload timeoutLarge videos — bump timeout=300 to e.g. 600 in the script

For Downstream Skills — MV3DT Export

A downstream Multi-View 3D Tracking skill fetches the MV3DT-format calibration directly from the microservice (this skill does not download it; it returns the project_id). After this skill reports COMPLETED:

  • GET /v1/result/{project_id}/mv3dt_result?result_type=amcmv3dt_output.zip (contains transforms.yml).
  • If VGGT ran to COMPLETED (Step 10): ?result_type=vggtvggt_mv3dt_output.zip.

Related Skills

  • skills/amc-setup-calibration-stack/SKILL.md — start MS + UI first.
  • skills/amc-run-sample-calibration/SKILL.md — verify the stack with the bundled sample before trying your own.
  • skills/amc-run-rtsp-calibration/SKILL.md — same calibration tail, but sourcing footage from live RTSP streams through VIOS.

Root README.md "Custom Dataset" and "Calibration Workflow (UI)" sections document input-video guidelines and the UI-driven alternative to this API flow.

<!-- signing marker -->

Frequently asked questions about Video Calibration

Similar skills