English

Revisiting Feature Alignment for One-stage Object Detection

Computer Vision and Pattern Recognition 2019-08-06 v1

Abstract

Recently, one-stage object detectors gain much attention due to their simplicity in practice. Its fully convolutional nature greatly reduces the difficulty of training and deployment compared with two-stage detectors which require NMS and sorting for the proposal stage. However, a fundamental issue lies in all one-stage detectors is the misalignment between anchor boxes and convolutional features, which significantly hinders the performance of one-stage detectors. In this work, we first reveal the deep connection between the widely used im2col operator and the RoIAlign operator. Guided by this illuminating observation, we propose a RoIConv operator which aligns the features and its corresponding anchors in one-stage detection in a principled way. We then design a fully convolutional AlignDet architecture which combines the flexibility of learned anchors and the preciseness of aligned features. Specifically, our AlignDet achieves a state-of-the-art mAP of 44.1 on the COCO test-dev with ResNeXt-101 backbone.

Keywords

Cite

@article{arxiv.1908.01570,
  title  = {Revisiting Feature Alignment for One-stage Object Detection},
  author = {Yuntao Chen and Chenxia Han and Naiyan Wang and Zhaoxiang Zhang},
  journal= {arXiv preprint arXiv:1908.01570},
  year   = {2019}
}
R2 v1 2026-06-23T10:39:40.525Z