Inside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks

@article{Bell2015InsideOutsideND,
  title={Inside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks},
  author={Sean Bell and C. Lawrence Zitnick and Kavita Bala and Ross B. Girshick},
  journal={2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2015},
  pages={2874-2883}
}
It is well known that contextual and multi-scale representations are important for accurate visual recognition. In this paper we present the Inside-Outside Net (ION), an object detector that exploits information both inside and outside the region of interest. Contextual information outside the region of interest is integrated using spatial recurrent neural networks. Inside, we use skip pooling to extract information at multiple scales and levels of abstraction. Through extensive experiments we… CONTINUE READING

Figures, Tables, Results, and Topics from this paper.

Key Quantitative Results

  • ION improves state-of-the-art on PASCAL VOC 2012 object detection from 73.9% to 77.9% mAP. On the new and more challenging MS COCO dataset, we improve state-of-the-art from 19.7% to 33.1% mAP.

Similar Papers

Citations

Publications citing this paper.
SHOWING 1-10 OF 363 CITATIONS

Small traffic sign detection from large image

  • Applied Intelligence
  • 2019
VIEW 4 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

MFR-CNN: Incorporating Multi-Scale Features and Global Information for Traffic Object Detection

  • IEEE Transactions on Vehicular Technology
  • 2018
VIEW 10 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Multi-scale Fusion with Context-aware Network for Object Detection

  • 2018 24th International Conference on Pattern Recognition (ICPR)
  • 2018
VIEW 8 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Object Detection with Deep Learning: A Review

  • IEEE transactions on neural networks and learning systems
  • 2018
VIEW 13 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Deep Learning Models for Context-Aware Object Detection

VIEW 13 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

YOLO9000: Better, Faster, Stronger

  • 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2016
VIEW 7 EXCERPTS
HIGHLY INFLUENCED

Object Detection Based on Region Decomposition and Assembly

VIEW 4 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Deep Learning for Generic Object Detection: A Survey

  • ArXiv
  • 2018
VIEW 8 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

FILTER CITATIONS BY YEAR

2015
2019

CITATION STATISTICS

  • 53 Highly Influenced Citations

  • Averaged 106 Citations per year from 2017 through 2019

References

Publications referenced by this paper.
SHOWING 1-10 OF 43 REFERENCES

Object Detection via a Multi-region and Semantic Segmentation-Aware CNN Model

  • 2015 IEEE International Conference on Computer Vision (ICCV)
  • 2015
VIEW 8 EXCERPTS
HIGHLY INFLUENTIAL

The Pascal Visual Object Classes (VOC) Challenge

VIEW 11 EXCERPTS
HIGHLY INFLUENTIAL

Fast R-CNN

  • 2015 IEEE International Conference on Computer Vision (ICCV)
  • 2015
VIEW 7 EXCERPTS
HIGHLY INFLUENTIAL

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

  • IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 2015
VIEW 5 EXCERPTS
HIGHLY INFLUENTIAL

Selective Search for Object Recognition

VIEW 6 EXCERPTS
HIGHLY INFLUENTIAL

What Makes for Effective Detection Proposals?

  • IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 2016
VIEW 1 EXCERPT
HIGHLY INFLUENTIAL