Unverified50% confidenceFactExact time
AI合成人脸缺乏真实血流，相应区域像素波动随机或近乎恒定，可作为换脸鉴定依据
1
Sources
50%
Confidence
Long-term
Relevance
7/18/2026
First Seen
Sources
Related Claims
Unverified人物一致性是AI视觉创作中公认的技术难点，因主流扩散模型每次生成引入随机性导致面部、体型、肤色漂移75% similarUnverifiedCurrent mainstream AI face-swapping models are prone to image stretching, twitching, or misalignment when encountering occlusion or large-angle profile views.74% similarUnverifiedAI深度伪造检测技术通过识别面部边缘频域异常、眨眼频率不自然、肤色与颈部不一致、背景与前景光照矛盾等伪造痕迹来检测换脸视频74% similarUnverifiedPhotorealistic portraits generated by GPT-Image-2 closely approximate reality in terms of skin texture, light-and-shadow relationships, and subtle facial expressions, to the point where even users with three years of AI image generation experience find it difficult to distinguish them from real photos.70% similarUnverifiedPixel art generated by AI is often not true pixel art; when enlarged, the pixel grid is uneven, and individual pixel blocks contain gradients, blurred edges, and inconsistent colors.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/551994API
curl https://kongchang.com/api/v1/knowledge/claims/551994MCP
get_claim(id=551994)