Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Object detection based on multiple trick feature pyramid networks and dynamic balanced L1 loss

Object detection based on multiple trick feature pyramid networks and dynamic balanced L1 loss Although the performance of the object detection has been significantly optimised in recent years, there is still a lot of room for designing multi-scale feature fusion methods and designing loss functions. Specifically, we propose Multiple Trick Feature Pyramid Networks (MT-FPN), by using various techniques such as feedback information, global module, attention mechanism, and fusion of refined information, to solve the problem of insufficient multi-scale feature fusion. We also propose Dynamic Balanced L1 Loss (DBLL), by utilising dynamic strategies and solving the derivative discontinuity problem, in order to help relieve the inconsistent problem between the dynamic training process and the fixed parameters. Moreover, by replacing FPN with MT-FPN, our Average Precision (AP) on Microsoft Common Objects in Context (MSCOCO) is 5.1 points and 3.8 points higher than FPN Faster R-CNN and Libra R-CNN, respectively. Without any bells and whistles, our experiments also show that the combined application of MT-FPN and DBLL achieves competitive performance compared with most advanced detectors on MS COCO benchmark. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Wireless and Mobile Computing Inderscience Publishers

Object detection based on multiple trick feature pyramid networks and dynamic balanced L1 loss

Download PDF
11 pages

Loading next page...
 
/lp/inderscience-publishers/object-detection-based-on-multiple-trick-feature-pyramid-networks-and-rx3P0ZMrtN

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Inderscience Publishers
Copyright
Copyright © Inderscience Enterprises Ltd
ISSN
1741-1084
eISSN
1741-1092
DOI
10.1504/ijwmc.2022.122489
Publisher site
See Article on Publisher Site

Abstract

Although the performance of the object detection has been significantly optimised in recent years, there is still a lot of room for designing multi-scale feature fusion methods and designing loss functions. Specifically, we propose Multiple Trick Feature Pyramid Networks (MT-FPN), by using various techniques such as feedback information, global module, attention mechanism, and fusion of refined information, to solve the problem of insufficient multi-scale feature fusion. We also propose Dynamic Balanced L1 Loss (DBLL), by utilising dynamic strategies and solving the derivative discontinuity problem, in order to help relieve the inconsistent problem between the dynamic training process and the fixed parameters. Moreover, by replacing FPN with MT-FPN, our Average Precision (AP) on Microsoft Common Objects in Context (MSCOCO) is 5.1 points and 3.8 points higher than FPN Faster R-CNN and Libra R-CNN, respectively. Without any bells and whistles, our experiments also show that the combined application of MT-FPN and DBLL achieves competitive performance compared with most advanced detectors on MS COCO benchmark.

Journal

International Journal of Wireless and Mobile Computing – Inderscience Publishers

Published: Jan 1, 2022

There are no references for this article.