AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Infant-ID: Fingerprints for Global Good

Joshua J. Engelsma, Debayan Deb, Kai Cao, Anjoo Bhatnagar, Prem Sewak Sudhish, Anil K. Jain

TPAMI 2021 · Architectures and efficiency › RNNs and normalization · approach: Learned embeddings for retrieval · 46 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add high-resolution minutiae extractor (ablation)Infant-Prints · Infant authentication, enrolled 0-1 monthsTAR @ FAR=0.1%: 35.3% -> 64.7%Improved
add high-resolution minutiae extractor (ablation)Infant-Prints · Infant authentication, enrolled 2-3 monthsTAR @ FAR=0.1%: 90.4% -> 92.8%Improved
add aging and enhancement (ablation)Infant-Prints · Infant authentication, enrolled 1-2 monthsTAR @ FAR=0.1%: 50.0% -> 63.9%Improved
add aging and enhancement (ablation)Infant-Prints · Infant authentication, enrolled 0-1 monthsTAR @ FAR=0.1%: 47.1% -> 64.7%Improved
add finetuning of DeepPrint (ablation)Infant-Prints · Infant authentication, enrolled 1-2 monthsTAR @ FAR=0.1%: 58.33% -> 63.9%Improved
add gender metadata (ablation)Infant-Prints · Infant authentication, enrolled 1-2 monthsTAR @ FAR=0.1%: 52.8% -> 63.9%Improved
add gender metadata (ablation)Infant-Prints · Infant authentication, enrolled 0-1 monthsTAR @ FAR=0.1%: 58.8% -> 64.7%Improved
use all improvements over preliminary algorithm (ablation)Infant-Prints · Infant authentication, enrolled 2-3 monthsTAR @ FAR=0.1%: 86.7% -> 92.8%Improved
fuse DeepPrint, Verifinger and Latent Matcher instead of best single matcherLatent Matcher · Infant authentication, enrolled 2-3 monthsTAR @ FAR=0.1%: 84.3% -> 92.8%Improved
fuse three matchersVerifinger · Infant search, enrolled 2-3 monthsRank-1 accuracy: 90.4% -> 90.4%No clear effect
use 1,900 ppi MSU RaspiReader instead of 500 ppi Digital Persona (ablation)Infant-Prints · Reader ablation, enrolled 2-3 monthsTAR @ FAR=0.1%: 52.1% -> 93.2%Improved
use 1,900 ppi MSU RaspiReader instead of 500 ppi Digital Persona (ablation)Infant-Prints · Reader ablation, enrolled 0-1 monthsTAR @ FAR=0.1%: 0% -> 58.3%Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy