AdGAP: Advanced Global Average Pooling

Arna Ghosh, Biswarup Bhattacharya, Somnath Basu Roy Chowdhury

AAAI 2018 pp. 8081-8082

doi:10.1609/AAAI.V32I1.12154 /aaai/2018/ghosh2018aaai-adgap/

Abstract

Global average pooling (GAP) has been used previously to generate class activation maps. The motivation behind AdGAP comes from the fact that the convolutional filters possess position information of the essential features and hence, combination of the feature maps could help us locate the class instances in an image. Our novel architecture generates promising results and unlike previous methods, the architecture is not sensitive to the size of the input image, thus promising wider application.

PDF AAAI Semantic Scholar

Cite

Text

Ghosh et al. "AdGAP: Advanced Global Average Pooling." AAAI Conference on Artificial Intelligence, 2018. doi:10.1609/AAAI.V32I1.12154

Markdown

[Ghosh et al. "AdGAP: Advanced Global Average Pooling." AAAI Conference on Artificial Intelligence, 2018.](https://mlanthology.org/aaai/2018/ghosh2018aaai-adgap/) doi:10.1609/AAAI.V32I1.12154

BibTeX

@inproceedings{ghosh2018aaai-adgap,
  title     = {{AdGAP: Advanced Global Average Pooling}},
  author    = {Ghosh, Arna and Bhattacharya, Biswarup and Chowdhury, Somnath Basu Roy},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2018},
  pages     = {8081-8082},
  doi       = {10.1609/AAAI.V32I1.12154},
  url       = {https://mlanthology.org/aaai/2018/ghosh2018aaai-adgap/}
}