SpecDETR: A Transformer-based Hyperspectral Point Object Detection Network

arXiv:2405.10148v1 Announce Type: new
Abstract: Hyperspectral target detection (HTD) aims to identify specific materials based on spectral information in hyperspectral imagery and can detect point targets, some of which occupy a smaller than one-pixel area. However, existing HTD methods are developed based on per-pixel binary classification, which limits the feature representation capability for point targets. In this paper, we rethink the hyperspectral point target detection from the object detection perspective, and focus more on the object-level prediction capability rather than …

abstract arxiv binary capability classification cs.cv detection feature however identify information materials network object per pixel representation targets transformer type