DGPE: A Dual-Stream Geometric-Aware Positional Encoding for Transformer-Based Object Detection

Accurately modeling complex spatial relationships is a pivotal challenge for Transformer-based object detectors, as existing models often rely on static positional encodings that fail to capture dynamic, real-world scene layouts. To address this issue, this work proposes a novel Dual-stream Geometri...

Full description

Bibliographic Details

Main Authors:	Han-Cheng Hsiang, Zunliang Yan, Yong Liu
Format:	Artigo
Language:	Inglês
Published:	IEEE 2025-01-01
Series:	IEEE Access
Subjects:	Attention mechanism object detection positional encoding spatial reasoning Transformer
Online Access:	https://ieeexplore.ieee.org/document/11278171/
Tags:	Add Tag No Tags, Be the first to tag this record!