logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

A multi-view feature collaborative optimization method for object detection

delete2026-05-16
delete0
PRE
AI
J
Jiaji Liu
R
Runhai Jiao *
X
Xiangning Zhan
K
Kaihang Li
DOI：10.1016/j.imavis.2026.106023delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• Develops a pseudo-multi-view feature learning framework to enhance object detection under single-view images. • Proposes an augmentation-based generation strategy to construct homogeneous multi-inputs. • Designs a global-local collaborative cosine loss to model cross-view consistency and complementarity. • Achieves consistent mAP gains across different benchmarks and detectors.
Keywords:
pseudo-multi-view
feature learning
object detection
cosine loss
augmentation strategy

Journal

Image and Vision Computing cover
Image and Vision Computing
IF:
4.2
Papers:
4.1K
Citations:
6.7K

Organization

No organization information available
Cited Papers

Cited Papers

No cited papers available