AI Paper Atlas

Research map / Federated learning and privacy / Privacy attacks and secure learning

Faceless Person Recognition: Privacy Implications in Social Media

Seong Joon Oh, Rodrigo Benenson, Mario Fritz, Bernt Schiele

ECCV 2016 · Federated learning and privacy › Privacy attacks and secure learning · approach: Visual privacy and face de-identification · 144 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
fine-tune pair matcher on tagged samples split0 (ablation)Siamese pair matching, visible · PIPA pairs within eventsEER (%): 79.1 -> 92.7Improved
fine-tune pair matcher on tagged samples split0 (ablation)Siamese pair matching, visible · PIPA pairs across eventsEER (%): 74.5 -> 81.4Improved
add pairwise terms with negative edge pruning (ablation)Faceless recognition CRF · PIPA val S2 black, across eventsaccuracy (%): 14.7 -> 39.2Improved
add pairwise terms with negative edge pruning (ablation)Faceless recognition CRF · PIPA val S2 black, within eventsaccuracy (%): 80.9 -> 81.9Improved
add pairwise terms without negative edge pruning (ablation)Faceless recognition CRF · PIPA val S2 black, within eventsaccuracy (%): 81.9 -> 47.3Hurt
add negative edge pruning to pairwise terms (ablation)Faceless recognition CRF · PIPA val S2 black, across eventsaccuracy (%): 29.4 -> 39.2Improved
adapt convnet to blurred heads (ablation)Unary person recogniser · PIPA val blur, across eventsaccuracy (%): 23.5 -> 28.8Improved
adapt convnet to blurred heads (ablation)Unary person recogniser · PIPA val blur, within eventsaccuracy (%): 84.3 -> 86.7Improved
adapt convnet to black fill-in (ablation)Unary person recogniser · PIPA val black, within eventsaccuracy (%): 80.1 -> 80.9No clear effect
use white instead of black fill-in obfuscationUnary person recogniser · PIPA val within events adaptedaccuracy (%): 80.9 -> 79.6No clear effect
increase tag rate from 1.25 to 10 instances per identityUnary+Pairwise · PIPA test visible, within eventsaccuracy (%): 56.8 -> 86.4Improved
add pairwise terms (ablation)Faceless recognition CRF · PIPA test S3 black, across eventsaccuracy (%): 19.6 -> 23.9Improved

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