@inproceedings{d9778d0a76754c35a66bd94bc4343f9e,
title = "BLNet: Boundary Points Localization Network for Object Detection",
abstract = "Currently, almost all of the two-stage object detectors treat bounding box localization as an offset regression problem in the second stage. However, the spatial information in each Region of Interest (RoI) feature map is not considered in this pipeline. In this paper, we propose a novel boundary points localization network (BLNet) to predict the location of four boundary points (topmost, bottommost, leftmost, rightmost) of objects on RoI feature maps with a fully convolutional network. In addition, in order to compensate for the low resolution of the heatmaps, we use a differentiable operation called soft-argmax to convert the heatmaps into the numerical coordinates directly. Experiments on PASCAL VOC 2007 and 2012 datasets demonstrate that our BLNet significantly outperforms the traditional regression-based methods. Using ResNet-101 as the backbone, our method achieves 80.9% mAP on VOC 2007 and 78.7% mAP on VOC 2012 dataset.",
keywords = "CNN, deep learning, object detection, object localization",
author = "Jiaoyang An and Bo Ma",
note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 5th IEEE International Conference on Signal and Image Processing, ICSIP 2020 ; Conference date: 23-10-2020 Through 25-10-2020",
year = "2020",
month = oct,
day = "23",
doi = "10.1109/ICSIP49896.2020.9339446",
language = "English",
series = "2020 IEEE 5th International Conference on Signal and Image Processing, ICSIP 2020",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "281--285",
booktitle = "2020 IEEE 5th International Conference on Signal and Image Processing, ICSIP 2020",
address = "United States",
}