deusyu/cosmetic-detect

Cosmetic Surgery Detection

Core Principle

Cosmetic procedures alter human tissue in ways that diverge from natural developmental patterns. Detection means identifying "anti-natural" signals — places where anatomy, proportion, texture, or dynamics break the statistical norms of unmodified faces/bodies.

This is adversarial: the best work is designed to be undetectable. Never claim certainty — use probability language ("consistent with," "suggestive of," "possible indicator of").

Analysis Protocol

Before analysis, read references/analysis-framework.md for the detailed region-by-region indicator checklist.

Step 1: Initial Assessment

• Image quality: Resolution, lighting, angle, makeup level. Low quality or heavy filters significantly reduce reliability — say so.
• Apparent ethnicity/ancestry: Establishes anatomical baseline. A "high nose bridge" is normal for Europeans but statistically unusual for East Asians.
• Apparent age: Sets expectations for skin quality, volume, aging signs.
• Filters/editing: Check for digital manipulation (smoothing, warping, face-tuning) — flag these as NOT cosmetic surgery to avoid false positives.

Step 2: Region-by-Region Analysis

Analyze each region independently using indicators from the reference file. For each region assess:

1. Are features within normal range for the person's apparent ethnicity and age?
2. Are there specific indicators of surgical or non-surgical intervention?
3. Confidence level: Low / Medium / High

Step 3: Cross-Region Coherence Check

The most powerful detection layer. Natural faces have internal consistency. Look for:

• Ethnic coherence: Do all features align with one consistent genetic background? (e.g., East Asian bone structure + Caucasian nose bridge = mismatch)
• Age coherence: Do all regions show consistent aging? (smooth forehead but aged hands = possible Botox)
• Symmetry: Natural faces have asymmetry. Excessive bilateral symmetry suggests correction.
• Proportion harmony: Do ratios between features fall within natural ranges?

Step 4: Output

## 整容检测分析 / Cosmetic Procedure Detection Analysis

### 基础信息 / Baseline
- 图像质量评估 / Image quality assessment
- 参考人种基线 / Ethnic baseline reference
- 年龄估计 / Estimated age
- 滤镜/修图评估 / Filter/editing assessment

### 区域分析 / Regional Analysis
For each region with findings:
- 观察到的特征 / Observed features
- 可能的项目 / Possible procedure(s)
- 置信度 / Confidence: Low|Medium|High
- 判断依据 / Reasoning

### 整体协调性 / Cross-Region Coherence
- 种族特征一致性 / Ethnic feature consistency
- 年龄一致性 / Age consistency
- 对称性分析 / Symmetry analysis

### 总评 / Overall Assessment
- 自然度评分 / Naturalness score (1-10, 10=completely natural)
- 最可能的项目清单 / Most likely procedures (if any)
- 整体置信度 / Overall confidence
- 重要声明 / Important disclaimer

Use the user's language. Template above is bilingual for reference.

Special Modes

Before/After Comparison

When 2+ photos of the same person at different times are provided:

• Align facial landmarks mentally between photos
• Prioritize skeletal/structural changes as highest confidence (bone/cartilage don't change naturally)
• Volume changes could be aging, weight, OR fillers
• Skin/texture changes could be aging, skincare, OR procedures

Celebrity/Public Figure

• Use knowledge of their appearance history if available
• Note that top-tier surgeons' work is hardest to detect
• Be especially careful with confidence levels

Batch Analysis

When analyzing multiple people (group photo, set of photos):

• Analyze each person independently
• Use the group as a natural baseline for comparison

Guidelines

• Never claim certainty. Even experienced surgeons can't always tell from photos.
• Acknowledge limitations. Lighting, angle, makeup, filters, genetics, image quality all affect analysis.
• Distinguish surgical vs non-surgical. Rhinoplasty vs Botox have different visual signatures — clearly separate them.
• Stay neutral. No judgment about whether someone "should" or "shouldn't" have had work done.
• Cultural sensitivity. Double eyelid surgery is extremely common in East Asia. Rhinoplasty is common globally. Note neutrally.