Researchers have developed InViStream, a novel system designed to preserve privacy in real-time volumetric video streaming. Unlike traditional video, volumetric video captures 3D scenes from multiple angles, making privacy redaction more complex. InViStream addresses this by removing private objects and content at the source before data is fused, preventing privacy leakage in the reconstructed 3D representation. The system combines object detection with depth-aware masking and ensures consistent privacy decisions across calibrated views, achieving high accuracy in both synthetic and real-world scenarios while maintaining real-time streaming speeds. AI
IMPACT This research could enable more secure and private immersive applications by addressing the unique privacy challenges of volumetric video.
RANK_REASON The cluster contains a research paper detailing a new system for privacy-preserving video streaming. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
