Saved in:
Bibliographic Details
Main Authors: Chen, Kean, Li, Jianguo, Lin, Weiyao, See, John, Wang, Ji, Duan, Lingyu, Chen, Zhibo, He, Changwei, Zou, Junni
Format: Preprint
Published: 2019
Subjects:
Online Access:https://arxiv.org/abs/1904.06373
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910035957776384
author Chen, Kean
Li, Jianguo
Lin, Weiyao
See, John
Wang, Ji
Duan, Lingyu
Chen, Zhibo
He, Changwei
Zou, Junni
author_facet Chen, Kean
Li, Jianguo
Lin, Weiyao
See, John
Wang, Ji
Duan, Lingyu
Chen, Zhibo
He, Changwei
Zou, Junni
contents One-stage object detectors are trained by optimizing classification-loss and localization-loss simultaneously, with the former suffering much from extreme foreground-background class imbalance issue due to the large number of anchors. This paper alleviates this issue by proposing a novel framework to replace the classification task in one-stage detectors with a ranking task, and adopting the Average-Precision loss (AP-loss) for the ranking problem. Due to its non-differentiability and non-convexity, the AP-loss cannot be optimized directly. For this purpose, we develop a novel optimization algorithm, which seamlessly combines the error-driven update scheme in perceptron learning and backpropagation algorithm in deep networks. We verify good convergence property of the proposed algorithm theoretically and empirically. Experimental results demonstrate notable performance improvement in state-of-the-art one-stage detectors based on AP-loss over different kinds of classification-losses on various benchmarks, without changing the network architectures. Code is available at https://github.com/cccorn/AP-loss.
format Preprint
id arxiv_https___arxiv_org_abs_1904_06373
institution arXiv
publishDate 2019
record_format arxiv
spellingShingle Towards Accurate One-Stage Object Detection with AP-Loss
Chen, Kean
Li, Jianguo
Lin, Weiyao
See, John
Wang, Ji
Duan, Lingyu
Chen, Zhibo
He, Changwei
Zou, Junni
Computer Vision and Pattern Recognition
One-stage object detectors are trained by optimizing classification-loss and localization-loss simultaneously, with the former suffering much from extreme foreground-background class imbalance issue due to the large number of anchors. This paper alleviates this issue by proposing a novel framework to replace the classification task in one-stage detectors with a ranking task, and adopting the Average-Precision loss (AP-loss) for the ranking problem. Due to its non-differentiability and non-convexity, the AP-loss cannot be optimized directly. For this purpose, we develop a novel optimization algorithm, which seamlessly combines the error-driven update scheme in perceptron learning and backpropagation algorithm in deep networks. We verify good convergence property of the proposed algorithm theoretically and empirically. Experimental results demonstrate notable performance improvement in state-of-the-art one-stage detectors based on AP-loss over different kinds of classification-losses on various benchmarks, without changing the network architectures. Code is available at https://github.com/cccorn/AP-loss.
title Towards Accurate One-Stage Object Detection with AP-Loss
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/1904.06373