Scale Pyramid Attention for Single Shot MultiBox Detector

Abstract

Feature pyramid mechanism has improved the performance of object detectors by a large margin, especially for the objects with small scale. As one of the first attempts to use pyramidal feature hierarchy, Single Shot MultiBox Detector (SSD) has largely accelerated the pipeline of detection with the competitive performance. Even if the feature pyramid mechanism is used for better detecting objects with small scale, the complex background information still misleads the network to focus on invalid areas. Image areas where small objects exist are more easily to be misjudged which influences the performance of lowerlevel layer in feature pyramid. In addition, a great deal of false positives may also be introduced into each pyramid layer because of the complex background. In this paper, we propose a novel method named scale pyramid attention hierarchy to better assist the SSD. The general feature pyramid detector can be guided to focus on the valid image areas as well as neglect the influence of complex background with the assist of this proposed module. Furthermore, the end-to-end training can highlight the foreground information which reduces the pressure of detectors, especially for detecting small objects. Flexible implementation makes it easily embedded in the feature pyramid mechanism. Experimental results on PASCAL VOC 2007, VOC 2012 and MS COCO confirm the effectiveness of the proposed module.

Maize (Zea mays L.) seedling detection based on the fusion of a modified deep learning model and a novel Lidar points projecting strategy

“…SSD is one representative of one-stage algorithms, its basic principle is to discretize the bounding box into the default bounding box in output space, then predict and adjust in the default bounding box [13] . During the predicting process, the network will give each object a score based on the object's existing situation, so as to well match the object's shape [13] . In addition, SSD has the unique character of multiscale feature mapping [13] .…”

“…During the predicting process, the network will give each object a score based on the object's existing situation, so as to well match the object's shape [13] . In addition, SSD has the unique character of multiscale feature mapping [13] .…”