Assisted Excitation of Activations: A Learning Technique to Improve Object Detectors

Mohammad Mahdi Derakhshani, Saeed Masoudnia, Amir Hossein Shaker, Omid Mersa, Mohammad Amin Sadeghi, Mohammad Rastegari, Babak N. Araabi; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 9201-9210

Abstract


We present a simple yet effective learning technique that significantly improves mAP of YOLO object detectors without compromising their speed. During network training, we carefully feed in localization information. We excite certain activations in order to help the network learn to better localize (Figure 2). In the later stages of training, we gradually reduce our assisted excitation to zero. We reached a new state-of-the-art in the speed-accuracy trade-off (Figure 1). Our technique improves the mAP of YOLOv2 by 3.8% and mAP of YOLOv3 by 2.2% on MSCOCO dataset. This technique is inspired from curriculum learning. It is simple and effective and it is applicable to most single-stage object detectors.

Related Material


[pdf]
[bibtex]
@InProceedings{Derakhshani_2019_CVPR,
author = {Derakhshani, Mohammad Mahdi and Masoudnia, Saeed and Shaker, Amir Hossein and Mersa, Omid and Sadeghi, Mohammad Amin and Rastegari, Mohammad and Araabi, Babak N.},
title = {Assisted Excitation of Activations: A Learning Technique to Improve Object Detectors},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}