Bottom-up modeling of repeated elements via single-image analysis-by-synthesis

Published in ECCV 2026, 2026

We propose a bottom-up, analysis-by-synthesis approach for discovering and modeling repeated elements from a single image, without relying on large-scale training data.

Project page: vayvi.github.io/repeated-elements

Recommended citation: Kalleli, S., Efros, A., Aubry, M. (2026). "Bottom-up modeling of repeated elements via single-image analysis-by-synthesis." ECCV.

Lien direct