Consistent Video Depth Estimation: Generating HQ Depth Maps From Single Video Input | Synced

Researchers from the University of Washington, Virginia Tech and Facebook have introduced an algorithm that can reconstruct dense, geometrically consistent depth for all pixels in monocular videos.

By Storm Warden · March 16, 2026 · 1 min read

ai
computer vision & graphics
machine learning & data science
research
3d reconstruction

Source: Synced | AI Technology & Industry Review

Researchers from the University of Washington, Virginia Tech and Facebook have introduced an algorithm that can reconstruct dense, geometrically consistent depth for all pixels in monocular videos.