Weighted Feature Pyramid Network for One-Stage Object Detection

One-stage object detection methods have attracted much attention for their high speed performance compared with two-stage methods. But one-stage methods under performs with small object detection. Feature Pyramid Network (FPN) was widely used to deal with this problem for its multi-scale feature present ability. However there still remains a few problems that are not considered in FPN, which results in limited improvement in detector performance. We note that FPN does not take the weight and scale distribution between different levels of feature maps into account when merging high-level feature maps and low-level feature maps. We present a network named Weighted Feature Pyramid Network (WFPN) to address these problems. Our experimental results on PASCAL VOC and MS COCO show that WFPN can significantly improve the detector performance, especially on small object detection.

[1]  Trevor Darrell,et al.  Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation , 2013, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[2]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[4]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[5]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Ali Farhadi,et al.  YOLOv3: An Incremental Improvement , 2018, ArXiv.

[7]  Ali Farhadi,et al.  YOLO9000: Better, Faster, Stronger , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[9]  Rogério Schmidt Feris,et al.  A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection , 2016, ECCV.

[10]  In-So Kweon,et al.  StairNet: Top-Down Semantic Aggregation for Accurate One Shot Detection , 2017, 2018 IEEE Winter Conference on Applications of Computer Vision (WACV).

[11]  Ross B. Girshick,et al.  Fast R-CNN , 2015, 1504.08083.

[12]  Shifeng Zhang,et al.  Single-Shot Refinement Neural Network for Object Detection , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[13]  Kaiming He,et al.  Feature Pyramid Networks for Object Detection , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  Rita Strack,et al.  Highly multiplexed imaging , 2015, Nature Methods.

[15]  Marios Savvides,et al.  Feature Selective Anchor-Free Module for Single-Shot Object Detection , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[16]  Ali Farhadi,et al.  You Only Look Once: Unified, Real-Time Object Detection , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[17]  Luc Van Gool,et al.  DeepProposals: Hunting Objects and Actions by Cascading Deep Convolutional Layers , 2016, International Journal of Computer Vision.

[18]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[19]  Yu-Wing Tai,et al.  Accurate Single Stage Detector Using Recurrent Rolling Convolution , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[20]  Wei Liu,et al.  SSD: Single Shot MultiBox Detector , 2015, ECCV.

[21]  Rachel Huang,et al.  YOLO-LITE: A Real-Time Object Detection Algorithm Optimized for Non-GPU Computers , 2018, 2018 IEEE International Conference on Big Data (Big Data).