False Match Rate (FMR)
Definition
The proportion of non-mate comparisons (different individuals) that the system incorrectly scores as a match. Also called the false positive rate or false accept rate. A lower FMR means fewer wrong identifications.
- Also called
- False positive rate, false accept rate
- Formula
- FMR = false matches / total non-mate comparisons
- Used in
- 1:1 verification and 1:N identification biometric systems
- Tradeoff partner
- False Non-Match Rate (FNMR), set by the match threshold
Common questions
How is FMR different from FNMR?+
FMR counts wrong matches between different people; FNMR counts missed matches between the same person. Lowering the match threshold to catch more true matches (cutting FNMR) almost always raises FMR, so the two are tuned together, not independently.
Why does FMR matter more in large-gallery searches?+
Even a very low per-pair FMR compounds across millions of comparisons in a 1:N search, so a system with an excellent per-pair rate can still return a meaningful number of false candidates when searched against a national database.
Related terms
- ArcFace / FaceNet
- Two influential deep CNN facial recognition architectures. FaceNet (Google, 2015) used a triplet-loss training approach to learn facial embeddings. ArcFace (Deng et...
- Embedding
- A compact numerical vector representation of a face produced by a neural network. The distance between two embeddings in the vector space...
- False Non-Match Rate (FNMR)
- The proportion of mate comparisons (same individual) that the system incorrectly scores as non-matching. Also called the false negative rate. A lower...
- Gallery vs Probe
- In facial recognition, the gallery is the reference database of enrolled face images; the probe is the query image being compared against...
- NIST FRVT
- The National Institute of Standards and Technology Face Recognition Vendor Testing programme, the primary independent benchmark for facial recognition algorithms, publishing ongoing...