CSPPNet: Cascade space pyramid pooling network for object detection

Ya‐Feng Liu

doi:10.1016/j.cviu.2025.104377

CSPPNet: Cascade space pyramid pooling network for object detection

2025-05-14

0

PRE

AI

DOI：10.1016/j.cviu.2025.104377

Original

Original request for help

Save

Abstract

En 中文

• We propose a Cascade Spatial Pyramid Pooling Network (CSPPNet) for real-time object detection. • We propose a cascade feature fusion (CFF) module. It adopts a three-branch cascade structure combined with GSConv to significantly optimize the feature extraction process. • We propose a nested spatial pooling (NSP) module. It effectively addresses the problem of detailed information loss in traditional max-pooling operations by integrating nested features through multi-scale pooling operations.

Keywords:

CSPPNet

cascade feature fusion

nested spatial pooling

real-time object detection

feature extraction

CSPPNet: Cascade space pyramid pooling network for object detection

Abstract

Journal

Organization