Cosine similarity

📖 Full definition

Cosine similarity computes the cosine of the angle between two vectors in a multi-dimensional space. The value ranges from -1 (opposite vectors) to +1 (identical vectors). For two facial embeddings, the closer to 1, the more similar the faces. This metric is insensitive to vector magnitude, making it especially suitable for normalised signature comparisons.

💡 Concrete example

If your face embedding and your father's give a cosine similarity of 0.78, you can convert to a resemblance percentage: 78%. For your mother, you might get 0.65 → 65%.

🔄 Synonyms

normalised dot product

🔗 See also

• Similarity score →