Login

Benchmarks

FVC-onGoing provides various benchmarks to evaluate and compare recognition algorithms. Each benchmark is based on a sequestered dataset that will not evolve over time; in case new datasets will be added in the future, they will form a different benchmark (or a new version of an existing one). Only results obtained on the same data will be compared.

Benchmarks are grouped into Benchmark Areas according to the (sub)problem addressed and the evaluation protocol adopted. In the following, the currently available benchmark areas and the corresponding benchmarks are briefly described.


Fingerprint VerificationBenchmarksLeaded by
This benchmark area contains fingerprint verification benchmarks. Fingerprint verification consists in comparing two fingerprints to determine whether they are impressions of the same finger or not (one-to-one comparisons). Algorithms submitted to these benchmarks are required to enroll fingerprints into proprietary or standard templates and to compare such templates to produce a similarity score. Read more...FV-STD-1.0
FV-TEST
FV-HARD-1.0
Palmprint VerificationBenchmarksLeaded by
This benchmark area contains palmprint verification benchmarks. Palmprint verification consists in comparing two palmprints to determine whether they are impressions of the same palm or not (one-to-one comparisons). Algorithms submitted to these benchmarks are required to enroll palmprints into proprietary or standard templates and to compare such templates to produce a similarity score. Read more...PV-TEST-FULL
PV-TEST-PARTIAL
PV-FULL-1.0
PV-PARTIAL-1.0
Fingerprint Matching (ISO)BenchmarksLeaded by
This benchmark area contains fingerprint matching benchmarks using a standard minutiae-based template format [ISO/IEC 19794-2 (2005)]. Algorithms submitted to these benchmarks are required to compare ISO fingerprint templates to determine whether they are impressions of the same finger or not (one-to-one comparisons). No fingerprint enrollment (feature extraction) is required, only the minutiae matching algorithms are evaluated by these benchmarks. Read more...FMISO-STD-1.0
FMISO-TEST
FMISO-HARD-1.0
Fingerprint IndexingBenchmarksLeaded by
This benchmark area contains fingerprint indexing benchmarks. Fingerprint indexing consists in comparing a query fingerprint against a large database and to select the most similar candidates. Algorithms submitted to these benchmarks are required to search a set of query fingerprints over a given fingerprint database, producing, for each query, a candidate list containing the most similar database fingerprints sorted by similarity. Read more...FIDX-TEST
FIDX-10K-1.0
FIDX-50K-1.0
Fingerprint Orientation ExtractionBenchmarksLeaded by
The estimation of local fingerprint orientations is a fundamental step in fingerprint analysis and recognition (e.g., it is a prerequisite for image enhancement). This benchmark area contains benchmarks for local orientation extraction algorithms. Algorithms submitted to these benchmarks are required to extract local orientations from fingerprint images and to save them into a specific format. The extracted orientations are compared to the ground-truth in order to assess the algorithm accuracy. Read more...FOE-TEST
FOE-STD-1.0
Secure Template Fingerprint VerificationBenchmarksLeaded by
This benchmark area contains fingerprint verification benchmarks for algorithms relying on protected templates to enhance privacy. Algorithms submitted to these benchmarks are required to enroll a given fingerprint into a protected template (that is, a template from which the fingerprint features cannot be extracted) and to compare it against a given fingerprint image. Read more...STFV-TEST
STFV-STD-1.0
STFV-HARD-1.0
Face Image ISO Compliance VerificationBenchmarksLeaded by
This benchmark area contains face image ISO compliance verification benchmarks. Algorithms submitted to these benchmarks are required to check the compliance of face images to ISO standard. Read more...FICV-TEST
FICV-1.0
Face Morphing ChallengeBenchmarksLeaded by
This benchmark area is aimed at analyzing the effects of image morphing on face recognition accuracy. The robustness against morphing alterations is evaluated by comparing morphed images against other images of the subjects used for the morphing. Algorithms submitted to these benchmarks are required to compare face images to determine whether they belong to the same subject or not (one-to-one comparisons). Read more...FMC-TEST
FMC-1.0
FMC-CRM-1.0
For information or suggestions: fvcongoing@csr.unibo.it Copyright © 2017 Biometric System Laboratory