公理愿景

设计与多媒体 社区

技能解读

当前原文的自动摘译,尚未人工审校。

快速参考

症状/任务 · 参考 对象分割、提升 · 参见 skills/vision-framework.md 手/身体姿势检测 · 参见 skills/vision-framework.md 文本识别 (OCR) · 参见 skills/vision-framework.md 条形码/QR 码检测 · 参见 skills/vision-framework.md

决策树

实施(姿势、分割、OCR、条形码、文档、实时扫描)? → skills/vision-framework.md 视觉智能系统集成(相机/截图搜索;iOS 26+、iPadOS27/macOS27)? → skills/vision-ref.md(视觉智能部分)点击分段、watchOS 上的视觉或基础模型的视觉工具(27 个周期)? → skills/vision-ref.md 需要 API 参考/代码示例? → skills/vision-ref.md

关键模式

实现 (skills/vision-framework.md):用于选择正确 Vision API 的决策树 使用 VisionKit 进行主题分割 隔离对象,同时排除手(组合 API) 手/身体姿势检测(21/19 地标)

反合理化

想法·现实“视觉框架只是一个请求/处理程序模式”·视觉具有坐标转换、置信阈值和性能陷阱。 Vision-framework.md 涵盖了它们。 “我无需技巧即可处理文本识别” · VNRecognizeTextRequest 具有快速/准确的模式和特定于语言的设置。 Vision-framework.md 有模式。 “主题分割很简单” · 实例蒙版具有 HDR 合成和手动排除模式。 Vision-framework.md 涵盖了复杂的场景。

调用示例

用户:“如何检测图像中的手部姿势?” → 请参阅skills/vision-framework.md 用户:“隔离主体但排除用户的手” → 请参阅skills/vision-framework.md 用户:“如何从图像中读取文本?”

流狐档案 运行与权限由流狐静态检查推断,不代表作者声明或实际测试结果。
领域
设计与多媒体
作者声明的智能体
未找到明确声明;不据此推断已兼容或已测试
静态检查
88 / 100 · 启发式扫描,不代表运行安全
运行环境证据
未执行;当前没有隔离运行测试证据
作者 / 版本 / 许可证
@CharlesWiltgen · 未声明许可证
上下文用量评级
低消耗
接入复杂程度
即装即用
是否需要外部接口密钥
不需要
兼容的系统
未声明(默认跨平台)
运行环境要求
未检出明确要求
检测到的文件与系统行为
  • 只读
网络行为
仅限本地
安装命令数
无(仅作为资料)

需要注意: 未声明工具权限范围;实际权限由运行环境决定。

作者说明文件(原文) SKILL.md


# Computer Vision

**You MUST use this skill for ANY computer vision work using the Vision framework.**

## Quick Reference

| Symptom / Task | Reference |
|----------------|-----------|
| Subject segmentation, lifting | See `skills/vision-framework.md` |
| Hand/body pose detection | See `skills/vision-framework.md` |
| Text recognition (OCR) | See `skills/vision-framework.md` |
| Barcode/QR code detection | See `skills/vision-framework.md` |
| Document scanning | See `skills/vision-framework.md` |
| DataScannerViewController | See `skills/vision-framework.md` |
| Structured document extraction (iOS 26+) | See `skills/vision-framework.md` |
| Isolate object excluding hand | See `skills/vision-framework.md` |
| Tap-to-segment any object `OS27` | See `skills/vision-ref.md` |
| Vision on watchOS `watchOS27` | See `skills/vision-ref.md` |
| Vision tools for Foundation Models (BarcodeReaderTool, OCRTool) `OS27` | See `skills/vision-ref.md` |
| Vision framework API reference | See `skills/vision-ref.md` |
| Visual Intelligence integration (iOS 26+, iPadOS27/macOS27) | See `skills/vision-ref.md` |
| Sensitive content classification (nudity/gore/violence), categorized via `detectedTypes` (`OS27`) | See `skills/vision-ref.md` |
| Group/cluster faces into people across a library, video highlights/key frames (`OS27`) | Use axiom-media (skills/media-intelligence.md) instead — MediaIntelligence clusters identities; Vision detects faces in one image |
| Subject not detected | See `skills/vision-diag.md` |
| Hand/body pose missing landmarks | See `skills/vision-diag.md` |
| Low confidence observations | See `skills/vision-diag.md` |
| UI freezing during processing | See `skills/vision-diag.md` |
| Coordinate conversion bugs | See `skills/vision-diag.md` |
| Text not recognized / wrong chars | See `skills/vision-diag.md` |
| Barcode not detected | See `skills/vision-diag.md` |
| DataScanner blank / no items | See `skills/vision-diag.md` |
| Document edges not detected | See `skills/vision-diag.md` |

## Decision Tree

```dot
digraph vision {
    start [label="Computer vision task" shape=ellipse];
    what [label="What do you need?" shape=diamond];

    start -> what;
    what -> "skills/vision-framework.md" [label="implement feature"];
    what -> "skills/vision-ref.md" [label="API reference"];
    what -> "skills/vision-ref.md" [label="Visual Intelligence"];
    what -> "skills/vision-ref.md" [label="tap-to-segment / watchOS / FM tools (27)"];
    what -> "skills/vision-diag.md" [label="something broken"];
}
```

1. Implementing (pose, segmentation, OCR, barcodes, documents, live scanning)? → `skills/vision-framework.md`
2. Visual Intelligence system integration (camera/screenshot search; iOS 26+, iPadOS27/macOS27)? → `skills/vision-ref.md` (Visual Intelligence section)
3. Tap-to-segment, Vision on watchOS, or Vision tools for Foundation Models (27 cycle)? → `skills/vision-ref.md`
4. Need API reference / code examples? → `skills/vision-ref.md`
5. Debugging issues (detection failures, confidence, coordinates)? → `skills/vision-diag.md`

## Critical Patterns

**Implementation** (`skills/vision-framework.md`):
- Decision tree for choosing the right Vision API
- Subject segmentation with VisionKit
- Isolating objects while excluding hands (combining APIs)
- Hand/body pose detection (21/19 landmarks)
- Text recognition (fast vs accurate modes)
- Barcode detection with symbology selection
- Document scanning and structured extraction (iOS 26+)
- Live scanning with DataScannerViewController
- CoreImage HDR compositing

**Diagnostics** (`skills/vision-diag.md`):
- Subject detection failures (edge of frame, lighting)
- Landmark tracking issues (confidence thresholds)
- Performance optimization (frame skipping, downscaling)
- Coordinate conversion (lower-left vs top-left origin)
- Text recognition failures (language, contrast)
- Barcode detection issues (symbology, size, glare)
- DataScanner troubleshooting (availability, data types)

## Anti-Rationalization

| Thought | Reality |
|---------|---------|
| "Vision framework is just a request/handler pattern" | Vision has coordinate conversion, confidence thresholds, and performance gotchas. vision-framework.md covers them. |
| "I'll handle text recognition without the skill" | VNRecognizeTextRequest has fast/accurate modes and language-specific settings. vision-framework.md has the patterns. |
| "Subject segmentation is straightforward" | Instance masks have HDR compositing and hand-exclusion patterns. vision-framework.md covers complex scenarios. |
| "Visual Intelligence is just the camera API" | Visual Intelligence is a system-level feature requiring IntentValueQuery and SemanticContentDescriptor. vision-ref.md has the integration section. |
| "I'll just process on the main thread" | Vision blocks UI on older devices. Users on iPhone 12 will experience frozen app. 15 min to add background queue. |

## Example Invocations

User: "How do I detect hand pose in an image?"
→ See `skills/vision-framework.md`

User: "Isolate a subject but exclude the user's hands"
→ See `skills/vision-framework.md`

User: "How do I read text from an image?"
→ See `skills/vision-framework.md`

User: "Scan QR codes with the camera"
→ See `skills/vision-framework.md`

User: "Subject detection isn't working"
→ See `skills/vision-diag.md`

User: "Text recognition returns wrong characters"
→ See `skills/vision-diag.md`

User: "Show me VNDetectHumanBodyPoseRequest examples"
→ See `skills/vision-ref.md`

User: "How do I make my app work with Visual Intelligence?"
→ See `skills/vision-ref.md`

User: "Let users tap an object in a photo to cut it out"
→ See `skills/vision-ref.md` (Iterative Segmentation)

User: "Can I use Vision in my watchOS app?"
→ See `skills/vision-ref.md` (Vision on watchOS)

User: "RecognizeDocumentsRequest API reference"
→ See `skills/vision-ref.md`

User: "Group faces into people across my library" / "cluster faces on-device into persons"
→ Use `axiom-media` (`skills/media-intelligence.md`) — identity clustering across assets, not per-image detection

讨论

基于 GitHub Discussions。登录 GitHub 即可参与讨论、点赞、订阅更新。