AI Paper Atlas

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

Towards a Visual Privacy Advisor: Understanding and Predicting Privacy Risks in Images

Tribhuvanesh Orekondy, Bernt Schiele, Mario Fritz

ICCV 2017 · Federated learning and privacy › Privacy attacks and secure learning · approach: Visual privacy and face de-identification · 177 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 the CNN end-to-end instead of training an SVM on frozen features (ResNet-50)privacy attribute prediction · VISPR testC-MAP: 40.50 -> 47.45Improved
fine-tune the CNN end-to-end instead of SVM (CaffeNet)privacy attribute prediction · VISPR testC-MAP: 37.93 -> 42.99Improved
fine-tune the CNN end-to-end instead of SVM (GoogleNet)privacy attribute prediction · VISPR testC-MAP: 39.88 -> 43.29Improved
use ResNet-50 instead of CaffeNet backbone (end-to-end)privacy attribute prediction · VISPR testC-MAP: 42.99 -> 47.45Improved
predict privacy risk directly with extra FC layers (PR-CNN) instead of from predicted attributes (AP-PR)privacy risk prediction · VISPR personalized riskL1-error: 0.656 -> 0.637Improved
PR-CNN instead of AP-PR at risk 4+privacy risk prediction · VISPR personalized risk (4+)MAP: 77.89 -> 78.38Improved
PR-CNN instead of AP-PR at risk 1+privacy risk prediction · VISPR personalized risk (1+)MAP: 94.94 -> 94.35Hurt
PR-CNN instead of AP-PR on safe imagesprivacy risk prediction · VISPR safe imagesL1-error: 0.34 -> 0.58Hurt
PR-CNN instead of AP-PR on private imagesprivacy risk prediction · VISPR private imagesL1-error: 0.84 -> 0.67Improved
use PR-CNN instead of users' own image-based judgmentprivacy risk estimation · Humans vs machineL1-error: 1.1 -> 1.03Improved
use AP-PR instead of users' own image-based judgmentprivacy risk estimation · Humans vs machineL1-error: 1.1 -> 1.27Hurt

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