Saved in:
Bibliographic Details
Main Authors: Yang, Guoyu, Wang, Yuan, Shi, Daming, Wang, Yanzhong
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.03325
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913720008966144
author Yang, Guoyu
Wang, Yuan
Shi, Daming
Wang, Yanzhong
author_facet Yang, Guoyu
Wang, Yuan
Shi, Daming
Wang, Yanzhong
contents Recent real-time semantic segmentation models, whether single-branch or multi-branch, achieve good performance and speed. However, their speed is limited by multi-path blocks, and some depend on high-performance teacher models for training. To overcome these issues, we propose Golden Cudgel Network (GCNet). Specifically, GCNet uses vertical multi-convolutions and horizontal multi-paths for training, which are reparameterized into a single convolution for inference, optimizing both performance and speed. This design allows GCNet to self-enlarge during training and self-contract during inference, effectively becoming a "teacher model" without needing external ones. Experimental results show that GCNet outperforms existing state-of-the-art models in terms of performance and speed on the Cityscapes, CamVid, and Pascal VOC 2012 datasets. The code is available at https://github.com/gyyang23/GCNet.
format Preprint
id arxiv_https___arxiv_org_abs_2503_03325
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Golden Cudgel Network for Real-Time Semantic Segmentation
Yang, Guoyu
Wang, Yuan
Shi, Daming
Wang, Yanzhong
Computer Vision and Pattern Recognition
Recent real-time semantic segmentation models, whether single-branch or multi-branch, achieve good performance and speed. However, their speed is limited by multi-path blocks, and some depend on high-performance teacher models for training. To overcome these issues, we propose Golden Cudgel Network (GCNet). Specifically, GCNet uses vertical multi-convolutions and horizontal multi-paths for training, which are reparameterized into a single convolution for inference, optimizing both performance and speed. This design allows GCNet to self-enlarge during training and self-contract during inference, effectively becoming a "teacher model" without needing external ones. Experimental results show that GCNet outperforms existing state-of-the-art models in terms of performance and speed on the Cityscapes, CamVid, and Pascal VOC 2012 datasets. The code is available at https://github.com/gyyang23/GCNet.
title Golden Cudgel Network for Real-Time Semantic Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.03325