Assisted Excitation of Activations: A Learning Technique to Improve Object Detectors
This work addresses the speed-accuracy trade-off for single-stage object detectors, offering an incremental improvement for computer vision applications.
The paper tackles the problem of improving object detection accuracy without sacrificing speed by introducing a learning technique that excites certain activations during training to enhance localization, resulting in a 3.8% mAP improvement for YOLOv2 and 2.2% for YOLOv3 on MSCOCO.
We present a simple and 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. 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. 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.