Building upon the SAM2 vision foundation model for downstream segmentation, this study introduces Boundary Enhanced Depth (BED)-SAM2. The SAM2 Hiera encoder architecture is modified to directly encode monocular depth information from RGB images, thereby providing geometric cues that enhance object boundary delineation and facilitate the extraction of camouflaged object shapes. BED-SAM2 demonstrates competitive state-of-the-art performance across multiple salient and camouflaged object detection tasks with as few as five training epochs.
@article{arxiv.2605.24893,
title = {BED-SAM2: Boundary-Enhanced-Depth SAM2 via Monocular Geometric Priors},
author = {Tyler Rust and Dara McNally and Kyle O'Donnell and Colin Kelly and Chandra Kambhamettu},
journal= {arXiv preprint arXiv:2605.24893},
year = {2026}
}
Comments
9 pages, 5 figures, 5 tables. Presented as a poster at the CVPR 2026 Workshop on Computer Vision in the Wild (CVinW). Code available at https://github.com/TylerRust-1/BED-SAM2