Methodology
Your photo is sent to an AI vision model with strict instructions: count faces, check angle, light and coverage, mark approximate positions of the features, and describe each feature using a fixed list of appearance words (for example “straight”, “gently curved”, “rounded tip”). The model cannot write free text into your report — we build every sentence from those fixed words.
Every “Traditional context” and “Explore the meaning” text comes from our own knowledge base of 26 reviewed English entries (version cf-kb-2026.10-v1), not from the AI. Entries are paraphrased by us from contemporary English-language overviews of Chinese physiognomy. Classical manuals such as the Ma Yi Shen Xiang (麻衣神相) and the Shen Xiang Quan Bian (神相全编) are the traditional reference works of the field; we name them, but we do not quote them, and nothing on this site is presented as a translated classical line.
See also the face reading chart and our privacy policy.