# Computer Vision Skill

> Computer Vision Skill is an Agent Skill for development published in the a5c-ai/babysitter repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Specialized skill for robot vision including feature detection, tracking, and camera calibration

- Skill name: `Computer Vision Skill`
- Category: [Development](https://www.openparable.com/skill/development/1)
- Source: https://github.com/a5c-ai/babysitter/blob/HEAD/library/specializations/robotics-simulation/skills/computer-vision/SKILL.md
- Repository: a5c-ai/babysitter (1,548 GitHub stars when collected)

## SKILL.md

# Computer Vision Skill

## Overview

Expert skill for robot vision applications including camera calibration, feature detection and tracking, stereo vision, and visual servoing.

## Capabilities

- Implement camera intrinsic calibration (pinhole, fisheye)
- Configure stereo camera calibration and rectification
- Set up camera-LiDAR extrinsic calibration
- Implement feature detection (ORB, SIFT, SURF, SuperPoint)
- Configure optical flow tracking (Lucas-Kanade, Farneback)
- Implement depth estimation from stereo
- Set up visual servoing pipelines
- Configure image undistortion and rectification
- Implement ArUco/AprilTag marker detection
- Set up hand-eye calibration

## Target Processes

- robot-calibration.js
- visual-slam-implementation.js
- object-detection-pipeline.js
- digital-twin-development.js

## Dependencies

- OpenCV
- cv_bridge
- image_geometry
- camera_calibration

## Usage Context

This skill is invoked when processes require camera calibration, feature detection, visual tracking, or image processing for robot vision applications.

## Output Artifacts

- Camera calibration files (YAML)
- Stereo calibration parameters
- Feature detection configurations
- Visual servoing controllers
- Image processing pipelines
- Marker detection configurations

---

Source: https://www.openparable.com/skill/development/computer-vision-skill (OpenParable). Cite the canonical URL above.
