a60a76b797
This snapshot establishes the camera-to-result recognition flow and related tests while checking in the project skill/docs assets required for the configured local tooling.
152 lines
5.2 KiB
Markdown
152 lines
5.2 KiB
Markdown
---
|
|
name: axiom-ios-vision
|
|
description: Use when implementing ANY computer vision feature - image analysis, object detection, pose detection, person segmentation, subject lifting, hand/body pose tracking.
|
|
license: MIT
|
|
---
|
|
|
|
# iOS Computer Vision Router
|
|
|
|
**You MUST use this skill for ANY computer vision work using the Vision framework.**
|
|
|
|
## When to Use
|
|
|
|
Use this router when:
|
|
- Analyzing images or video
|
|
- Detecting objects, faces, or people
|
|
- Tracking hand or body pose
|
|
- Segmenting people or subjects
|
|
- Lifting subjects from backgrounds
|
|
- Recognizing text in images (OCR)
|
|
- Detecting barcodes or QR codes
|
|
- Scanning documents
|
|
- Using VisionKit or DataScannerViewController
|
|
- Integrating with Visual Intelligence (iOS 26+ system camera feature)
|
|
|
|
## Routing Logic
|
|
|
|
### Vision Work
|
|
|
|
**Implementation patterns** → `/skill axiom-vision`
|
|
- Subject segmentation (VisionKit)
|
|
- Hand pose detection (21 landmarks)
|
|
- Body pose detection (2D/3D)
|
|
- Person segmentation
|
|
- Face detection
|
|
- Isolating objects while excluding hands
|
|
- Text recognition (VNRecognizeTextRequest)
|
|
- Barcode/QR detection (VNDetectBarcodesRequest)
|
|
- Document scanning (VNDocumentCameraViewController)
|
|
- Live scanning (DataScannerViewController)
|
|
- Structured document extraction (RecognizeDocumentsRequest, iOS 26+)
|
|
|
|
**API reference** → `/skill axiom-vision-ref`
|
|
- Complete Vision framework API
|
|
- VNDetectHumanHandPoseRequest
|
|
- VNDetectHumanBodyPoseRequest
|
|
- VNGenerateForegroundInstanceMaskRequest
|
|
- VNRecognizeTextRequest (fast/accurate modes)
|
|
- VNDetectBarcodesRequest (symbologies)
|
|
- DataScannerViewController delegates
|
|
- RecognizeDocumentsRequest (iOS 26+)
|
|
- Coordinate conversion patterns
|
|
|
|
**Visual Intelligence integration** → `/skill axiom-vision-ref` (see Visual Intelligence Integration section)
|
|
- Making app content discoverable to Visual Intelligence camera
|
|
- `IntentValueQuery` and `SemanticContentDescriptor`
|
|
- Deep linking from Visual Intelligence results
|
|
|
|
**Diagnostics** → `/skill axiom-vision-diag`
|
|
- Subject not detected
|
|
- Hand pose missing landmarks
|
|
- Low confidence observations
|
|
- Performance issues
|
|
- Coordinate conversion bugs
|
|
- Text not recognized or wrong characters
|
|
- Barcodes not detected
|
|
- DataScanner showing blank or no items
|
|
- Document edges not detected
|
|
|
|
## Decision Tree
|
|
|
|
1. Implementing (pose, segmentation, OCR, barcodes, documents, live scanning)? → vision
|
|
2. Visual Intelligence system integration (camera feature, iOS 26+)? → vision-ref (Visual Intelligence section)
|
|
3. Need API reference / code examples? → vision-ref
|
|
4. Debugging issues (detection failures, confidence, coordinates)? → vision-diag
|
|
|
|
## Anti-Rationalization
|
|
|
|
| Thought | Reality |
|
|
|---------|---------|
|
|
| "Vision framework is just a request/handler pattern" | Vision has coordinate conversion, confidence thresholds, and performance gotchas. vision covers them. |
|
|
| "I'll handle text recognition without the skill" | VNRecognizeTextRequest has fast/accurate modes and language-specific settings. vision has the patterns. |
|
|
| "Subject segmentation is straightforward" | Instance masks have HDR compositing and hand-exclusion patterns. vision covers complex scenarios. |
|
|
| "Visual Intelligence is just the camera API" | Visual Intelligence is a system-level feature requiring IntentValueQuery and SemanticContentDescriptor. vision-ref has the integration section. |
|
|
|
|
## Critical Patterns
|
|
|
|
**vision**:
|
|
- Subject segmentation with VisionKit
|
|
- Hand pose detection (21 landmarks)
|
|
- Body pose detection (2D/3D, up to 4 people)
|
|
- Isolating objects while excluding hands
|
|
- CoreImage HDR compositing
|
|
- Text recognition (fast vs accurate modes)
|
|
- Barcode detection (symbology selection)
|
|
- Document scanning with perspective correction
|
|
- Live scanning with DataScannerViewController
|
|
- Structured document extraction (iOS 26+)
|
|
|
|
**vision-diag**:
|
|
- Subject detection failures
|
|
- Landmark tracking issues
|
|
- Performance optimization
|
|
- Observation confidence thresholds
|
|
- Text recognition failures (language, contrast)
|
|
- Barcode detection issues (symbology, distance)
|
|
- DataScanner troubleshooting
|
|
- Document edge detection problems
|
|
|
|
## Example Invocations
|
|
|
|
User: "How do I detect hand pose in an image?"
|
|
→ Invoke: `/skill axiom-vision`
|
|
|
|
User: "Isolate a subject but exclude the user's hands"
|
|
→ Invoke: `/skill axiom-vision`
|
|
|
|
User: "How do I read text from an image?"
|
|
→ Invoke: `/skill axiom-vision`
|
|
|
|
User: "Scan QR codes with the camera"
|
|
→ Invoke: `/skill axiom-vision`
|
|
|
|
User: "How do I implement document scanning?"
|
|
→ Invoke: `/skill axiom-vision`
|
|
|
|
User: "Use DataScannerViewController for live text"
|
|
→ Invoke: `/skill axiom-vision`
|
|
|
|
User: "Subject detection isn't working"
|
|
→ Invoke: `/skill axiom-vision-diag`
|
|
|
|
User: "Text recognition returns wrong characters"
|
|
→ Invoke: `/skill axiom-vision-diag`
|
|
|
|
User: "Barcode not being detected"
|
|
→ Invoke: `/skill axiom-vision-diag`
|
|
|
|
User: "Show me VNDetectHumanBodyPoseRequest examples"
|
|
→ Invoke: `/skill axiom-vision-ref`
|
|
|
|
User: "What symbologies does VNDetectBarcodesRequest support?"
|
|
→ Invoke: `/skill axiom-vision-ref`
|
|
|
|
User: "RecognizeDocumentsRequest API reference"
|
|
→ Invoke: `/skill axiom-vision-ref`
|
|
|
|
User: "How do I make my app work with Visual Intelligence?"
|
|
→ Invoke: `/skill axiom-vision-ref`
|
|
|
|
User: "How do users discover my app content through the camera?"
|
|
→ Invoke: `/skill axiom-vision-ref`
|