TY - GEN

T1 - Feature pyramid networks for object detection

AU - Lin, Tsung Yi

AU - Dollár, Piotr

AU - Girshick, Ross

AU - He, Kaiming

AU - Hariharan, Bharath

AU - Belongie, Serge

N1 - Publisher Copyright: © 2017 IEEE.

PY - 2017/11/6

Y1 - 2017/11/6

N2 - Feature pyramids are a basic component in recognition systems for detecting objects at different scales. But recent deep learning object detectors have avoided pyramid representations, in part because they are compute and memory intensive. In this paper, we exploit the inherent multi-scale, pyramidal hierarchy of deep convolutional networks to construct feature pyramids with marginal extra cost. A top-down architecture with lateral connections is developed for building high-level semantic feature maps at all scales. This architecture, called a Feature Pyramid Network (FPN), shows significant improvement as a generic feature extractor in several applications. Using FPN in a basic Faster R-CNN system, our method achieves state-of-the-art single-model results on the COCO detection benchmark without bells and whistles, surpassing all existing single-model entries including those from the COCO 2016 challenge winners. In addition, our method can run at 5 FPS on a GPU and thus is a practical and accurate solution to multi-scale object detection. Code will be made publicly available.

AB - Feature pyramids are a basic component in recognition systems for detecting objects at different scales. But recent deep learning object detectors have avoided pyramid representations, in part because they are compute and memory intensive. In this paper, we exploit the inherent multi-scale, pyramidal hierarchy of deep convolutional networks to construct feature pyramids with marginal extra cost. A top-down architecture with lateral connections is developed for building high-level semantic feature maps at all scales. This architecture, called a Feature Pyramid Network (FPN), shows significant improvement as a generic feature extractor in several applications. Using FPN in a basic Faster R-CNN system, our method achieves state-of-the-art single-model results on the COCO detection benchmark without bells and whistles, surpassing all existing single-model entries including those from the COCO 2016 challenge winners. In addition, our method can run at 5 FPS on a GPU and thus is a practical and accurate solution to multi-scale object detection. Code will be made publicly available.

UR - http://www.scopus.com/inward/record.url?scp=85041898381&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2017.106

DO - 10.1109/CVPR.2017.106

M3 - Conference article

AN - SCOPUS:85041898381

SP - 936

EP - 944

JO - Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017

JF - Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017

T2 - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017

Y2 - 21 July 2017 through 26 July 2017

ER -