SCOUT: frozen-encoder embedding-space prediction for sim-to-real text-based person retrieval. ECCV 2026 Workshop (AI City Challenge Track 4), accepted as a poster.
-
Updated
Aug 2, 2026 - Python
SCOUT: frozen-encoder embedding-space prediction for sim-to-real text-based person retrieval. ECCV 2026 Workshop (AI City Challenge Track 4), accepted as a poster.
Frozen HSL features as a DDPM conditioning substrate. Reproducible feasibility run with matched controls and retained negative results; not SOTA.
Add a description, image, and links to the frozen-features topic page so that developers can more easily learn about it.
To associate your repository with the frozen-features topic, visit your repo's landing page and select "manage topics."