Small Object Detection using Context and Attention
Authors:
- 13 April 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 181-186
Abstract
There are many limitations applying object detection algorithm on various environments. Specifically, detecting small objects is still challenging because they have low-resolution and limited information. We propose an object detection method using context for improving accuracy of detecting small objects. The proposed method uses additional features from different layers as context by concatenating multi-scale features. We also propose object detection with attention mechanism which can focus on the object in image, and it can include contextual information from target layer. Experimental results shows that proposed method also has higher accuracy than conventional SSD on detecting small objects. Moreover, for $300 \times 300$ input, we achieved 78.1% Mean Average Precision (mAP) on the PASCAL VOC2007 test set.
Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- Perceptual Generative Adversarial Networks for Small Object Detection2017
- YOLO9000: Better, Faster, Stronger2017
- Residual Attention Network for Image Classification2017
- Enhancement of SSD by concatenating feature maps for object detection2017
- Deep Residual Learning for Image Recognition2016
- Fast R-CNN2015
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation2014
- Selective Search for Object RecognitionInternational Journal of Computer Vision, 2013
- The Pascal Visual Object Classes (VOC) ChallengeInternational Journal of Computer Vision, 2009