Human Biometric Analysis
Finger, Eye & Skin Detection
Uses MediaPipe to analyze human anatomy for incorrect finger counts, asymmetric eyes, unnatural skin texture, and other anatomical anomalies common in AI-generated faces.
使い方
AI image generators frequently struggle with human anatomy, especially hands and faces. Using MediaPipe machine learning, this detector counts fingers, measures eye symmetry, analyzes facial proportions, and examines skin texture patterns.
Fingers
Count all digits
Eyes
Check symmetry
Face
Analyze proportions
Skin
Texture analysis
Frequently Asked Questions
Why do AI images have wrong finger counts?
AI models learn hand appearance from training data without understanding that humans have exactly 5 fingers. They often generate 4, 6, or merged fingers because hands are complex and variable in photos.
How does eye symmetry detection work?
MediaPipe detects 468 facial landmarks and measures the ratio between eye sizes, positions, and angles. Real faces have natural asymmetry within specific ranges, while AI often creates unnaturally symmetric or asymmetric eyes.
Can this detect AI-generated portraits?
Yes, especially portraits with visible hands or multiple people. Even high-quality AI portraits often have subtle facial proportion errors or unnatural skin smoothness that this detector identifies.
What skin texture anomalies are detected?
The detector analyzes pore patterns, wrinkle consistency, and texture frequency. AI skin often lacks natural pore variation, has overly smooth areas, or shows repetitive texture patterns not found in real skin.
Does this work on cartoon-style images?
Biometric analysis is most effective on photorealistic images. Cartoon or stylized art intentionally deviates from human anatomy, so other detection methods are more suitable for those styles.
What facial proportions are checked?
The system measures eye-to-nose ratio, nose-to-mouth distance, face width proportions, and ear placement. AI often generates faces that look good at first glance but have proportions outside natural human ranges.
Is MediaPipe required for detection?
MediaPipe provides the best accuracy, but the system includes fallback methods using OpenCV Haar cascades when MediaPipe is unavailable. The fallback has reduced accuracy but still detects major anomalies.
How effective is this against DALL-E 3?
DALL-E 3 has significantly improved hand generation, but still makes errors in complex poses or multiple hands. The detector catches these remaining issues, though accuracy is lower than with older models.
What happens if no humans are in the image?
If no humans, hands, or faces are detected, this method returns a neutral score (0.5) and other detection methods carry more weight. The ensemble system automatically adjusts for content type.
How fast is biometric analysis?
MediaPipe is optimized for real-time processing. Full biometric analysis typically completes in 200-500ms on CPU, making it suitable for production use without significant latency.
関連方法
ML検出
数百万の画像で訓練された最先端のTransformerモデルを使用して、本物の写真とAI生成コンテンツを区別します。
PRNU分析
Photo Response Non-Uniformity (PRNU) detects unique camera sensor fingerprints from manufacturing imperfections. AI images cannot replicate these authentic sensor signatures.
周波数分析
DCT（離散コサイン変換）を用いて画像の高周波・低周波成分の分布を分析。AI生成画像はカメラで撮影された写真に存在する自然な高周波ノイズが欠如しており、この特徴で真偽を判定します。無料オンラインツール。
勾配分析
Analyzes edge patterns and texture characteristics using Sobel, Canny, and Laplacian operators. AI images often have unnaturally smooth or uniform gradients.
ノイズパターン
Real photographs contain unique noise patterns from camera sensors that vary across the image. AI-generated images have unnaturally uniform noise distribution.
メタデータ分析
Image metadata contains valuable clues about its origin. We analyze EXIF data, software signatures, and other embedded information to identify AI generation tools.
GANフィンガープリント
GAN（敵対的生成ネットワーク）が生成する画像のチェッカーボードパターン、カラーバンディング、スペクトル異常などの固有アーティファクトを高精度で検出。StyleGAN、ProGAN、CycleGAN対応の無料オンライン分析ツール。
テクスチャ分析
AI生成画像に見られるテクスチャ異常のLocal Binary Pattern分析。均一性、エントロピー、均質性を測定。
Anatomy Detection
AI image generators often create anatomical errors that humans immediately recognize as wrong. We use computer vision to detect these telltale mistakes.
C2PA Verification
C2PA (Coalition for Content Provenance and Authenticity) is an industry standard for tracking the origin and history of digital content through cryptographic signatures.
Semantic Inconsistency Detection
Detects logical inconsistencies like incorrect shadows, impossible perspectives, distorted reflections, and violations of physical laws that AI often produces.
Lighting Physics Validation
Validates light source consistency, shadow direction physics, specular highlight accuracy, and color temperature uniformity across the image.
Compression Artifact Analysis
Analyzes JPEG compression artifacts to estimate quality levels and detect re-compression patterns that indicate image manipulation or AI generation.
Edge Sharpness Analysis
Analyzes sharpness distribution across the image and validates depth-of-field consistency. AI often produces unnaturally uniform sharpness.
Statistical Pattern Analysis
Analyzes statistical properties including Shannon entropy, histogram patterns, and Benford's Law compliance to detect synthetic image characteristics.
Chromatic Aberration Analysis
Detects the absence of chromatic aberration (color fringing) that real camera lenses produce. AI images lack these optical artifacts.
Micro-Texture Analysis
Analyzes microscopic texture patterns for repetition, uniformity, and unnatural randomness that AI generators often exhibit.
Color Palette Analysis
Analyzes color distribution including saturation levels, color diversity, and white balance consistency. AI images often have oversaturated colors.
画像をチェック
All methods are combined using weighted scoring to produce a final verdict with confidence level.
今すぐ試す