Unverified50% confidenceFactExact time
gFD (generative Fréchet distance) is a metric for evaluating generated image quality, where a lower value indicates higher generated image quality.
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedFID (Fréchet Inception Distance) is a quantitative metric used to measure the distance between the distribution of generated images and real photographs.77% similarVerifiedFID measures the distributional distance between generated images and real images in feature space; a lower value indicates higher generation quality.76% similarUnverifiedEvaluating image generation models requires a multi-dimensional metric system including FID score, CLIP score, and inference speed.69% similarUnverifiedFID 由 Heusel 等人于2017年提出，通过比较真实图像与生成图像在 Inception 网络特征空间中的分布距离来衡量生成质量，数值越低越接近真实图像分布66% similarUnverifiedCFG scale（无分类器引导强度）控制生成图像对prompt的遵循程度65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073435API
curl https://kongchang.com/api/v1/knowledge/claims/1073435MCP
get_claim(id=1073435)