.png){kind=icon}
Return
RectyNet: Geometry-Aware Feature Rectification for Robust Object Detection
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patrec.2026.09.010.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Created a grocery dataset with 1946 images and 176 product classes.
• Introduced angle correction for rectifying skewed grocery images.
• Achieved [email protected] of 61.2 on angled images, outperforming SOTA models.
• Provides analysis of angle and density effects on detection accuracy.
• Proposed RectyNet for improved detection on angled and frontal images.
Keywords:
Product Identification
Object Detection
Image Rectification
Grocery Dataset
Image Computer Vision
Journal
IF:
3.3
Papers:
8.0K
Citations:
1.6W
Organization
No organization information available
Cited Papers
No cited papers available