arrow
返回

Instance Segmentation Based on Improved Self-Adaptive Normalization

delete2022-06-10
delete4
delete
OA
AI
杨
杨森 (Sen Yang)
X
Xiaobao Wang
Q
Qijuan Yang
E
Enzeng Dong
S
Shengzhi Du *
DOI:10.3390/s22124396delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
The single batch normalization (BN) method is commonly used in the instance segmentation algorithms. The batch size is concerned with some drawbacks. A too small sample batch size leads to a sharp drop in accuracy, but a too large batch may result in the memory overflow of graphic processing units (GPU). These problems make BN not feasible to some instance segmentation tasks with inappropriate batch sizes. The self-adaptive normalization (SN) method, with an adaptive weight loss layer, shows good performance in instance segmentation algorithms, such as the YOLACT. However, the parameter averaging mechanism in the SN method is prone to problems in the weight learning and assignment process. In response to such a problem, the paper proposes to replace the single BN with an adaptive weight loss layer in SN models, based on which a weight learning method is developed. The proposed method increases the input feature expression ability of the subsequent layers. By building a Pytorch deep learning framework, the proposed method is validated in the MS-COCO data set and Autonomous Driving Cityscapes data set. The experimental results prove that the proposed method is effective in processing samples independent from the batch size. The stable accuracy for all kinds of target segmentation is achieved, and the overall loss value is significantly reduced at the same time. The convergence speed of the network is also improved.
Keyword:
instance segmentation
normalized
batch size
self-adaptive normalization
adaptive weight
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

Sensors 封面图
Sensors
IF:
3.5
论文数:
7.2W
被引数:
20.9W

机构

T
Tshwane University of Technology
学者数:
2.2K
论文数: 1.8K
被引数: 2.4K
T
Tianjin University of Technology
学者数:
8.8K
论文数: 5.9K
被引数: 1.0W
引用论文

引用论文

SSN: Learning Sparse Switchable Normalization via SparsestMax
err2019-12-09
err13
PREAI
errShao, Wenqi; Li, Jingyu; Ren, Jiamin; Zhang, Ruimao; Wang, Xiaogang; Luo, Ping
err分享
err收藏
Resolution of Holliday junctions by RuvABC prevents dimer formation in rep mutants and UV‐irradiated cells
err2002-01-18
err0
errOAAI
errBénédicte Michel; Gavin D. Recchia; Marion Penel‐Colin; S. Dusko Ehrlich; David J. Sherratt
err分享
err收藏
学者 查看更多内容