Abstract
Convolutional Neural Networks (CNNs) use pooling to decrease the size of activation maps. This process is crucial to increase the receptive fields and to reduce computational requirements of subsequent convolutions. An important feature of the pooling operation is the minimization of information loss, with respect to the initial activation maps, without a significant impact on the computation and memory overhead. To meet these requirements, we propose SoftPool: a fast and efficient method for exponentially weighted activation downsampling. Through experiments across a range of architectures and pooling methods, we demonstrate that SoftPool can retain more information in the reduced activation maps. This refined downsampling leads to improvements in a CNN's classification accuracy. Experiments with pooling layer substitutions on ImageNet1K show an increase in accuracy over both original architectures and other pooling methods. We also test SoftPool on video datasets for action recognition. Again, through the direct replacement of pooling layers, we observe consistent performance improvements while computational loads and memory requirements remain limited.
| Original language | English |
|---|---|
| Title of host publication | Proceedings - 2021 IEEE/CVF International Conference on Computer Vision, ICCV 2021 |
| Publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| Pages | 10337-10346 |
| Number of pages | 10 |
| ISBN (Electronic) | 9781665428125 |
| DOIs | |
| Publication status | Published - 28 Feb 2022 |
| Event | 18th IEEE/CVF International Conference on Computer Vision, ICCV 2021 - Virtual, Online, Canada Duration: 11 Oct 2021 → 17 Oct 2021 |
Publication series
| Name | Proceedings of the IEEE International Conference on Computer Vision |
|---|---|
| ISSN (Print) | 1550-5499 |
| ISSN (Electronic) | 2380-7504 |
Conference
| Conference | 18th IEEE/CVF International Conference on Computer Vision, ICCV 2021 |
|---|---|
| Country/Territory | Canada |
| City | Virtual, Online |
| Period | 11/10/21 → 17/10/21 |
Bibliographical note
Publisher Copyright:© 2021 IEEE