New VIPA Attack Projects Adversarial Patterns to Fool Image Classifiers in Real Scenes
Researchers introduced VIPA, a view-invariant projector-based adversarial attack that physically manipulates real-world scenes by projecting adversarial patterns to fool deep image classifiers.
Read the full story at arXiv cs.CV ↗
Timeline · 1 report
- 2026-11-06 05:00 · arXiv cs.CV
VIPA: View-Invariant Projector-Based Adversarial Attack