PrivAR applies VLMs with chain-of-thought prompting to detect and mitigate semantic privacy risks in AR, reporting 81.48% accuracy and 17.58% leakage rate on a real-world dataset.
Beyond blanket masking: Examin- ing granularity for privacy protection in images captured by blind and low vision users,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
See No Evil: Semantic Context-Aware Privacy Risk Detection for AR
PrivAR applies VLMs with chain-of-thought prompting to detect and mitigate semantic privacy risks in AR, reporting 81.48% accuracy and 17.58% leakage rate on a real-world dataset.