# Refining activation downsampling with SoftPool

arXiv: https://arxiv.org/abs/2101.00440 · HTML: https://mlsanity.com/paper/2101.00440/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- Builds on 37 of its 70 references that are in the map.
- First posted on arXiv 2 Jan 2021.

## Builds on (most cited first, 30 of 37)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [Wide Residual Networks](https://mlsanity.com/paper/1605.07146/) (2016)
- [Fast R-CNN](https://mlsanity.com/paper/1504.08083/) (2015)
- [Aggregated Residual Transformations for Deep Neural Networks](https://mlsanity.com/paper/1611.05431/) (2016)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [SlowFast Networks for Video Recognition](https://mlsanity.com/paper/1812.03982/) (2018)
- [A Closer Look at Spatiotemporal Convolutions for Action Recognition](https://mlsanity.com/paper/1711.11248/) (2017)
- [Regularized Evolution for Image Classifier Architecture Search](https://mlsanity.com/paper/1802.01548/) (2018)
- [Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition](https://mlsanity.com/paper/1406.4729/) (2014)
- [Network In Network](https://mlsanity.com/paper/1312.4400/) (2013)
- [Sketch-based Manga Retrieval using Manga109 Dataset](https://mlsanity.com/paper/1510.04389/) (2015)
- [A Short Note on the Kinetics-700 Human Action Dataset](https://mlsanity.com/paper/1907.06987/) (2019)
- [Video Classification with Channel-Separated Convolutional Networks](https://mlsanity.com/paper/1904.02811/) (2019)
- [Inverting Visual Representations with Convolutional Networks](https://mlsanity.com/paper/1506.02753/) (2015)
- [Large-scale weakly-supervised pre-training for video action recognition](https://mlsanity.com/paper/1905.00561/) (2019)
- [Dual Path Networks](https://mlsanity.com/paper/1707.01629/) (2017)
- [Learning Spatio-Temporal Representation with Local and Global Diffusion](https://mlsanity.com/paper/1906.05571/) (2019)
- [A Multigrid Method for Efficiently Training Video Models](https://mlsanity.com/paper/1912.00998/) (2019)
- [Stochastic Pooling for Regularization of Deep Convolutional Neural Networks](https://mlsanity.com/paper/1301.3557/) (2013)
- [Generalizing Pooling Functions in Convolutional Neural Networks: Mixed, Gated, and Tree](https://mlsanity.com/paper/1509.08985/) (2015)
- [Multi-Fiber Networks for Video Recognition](https://mlsanity.com/paper/1807.11195/) (2018)
- [Learned-Norm Pooling for Deep Feedforward and Recurrent Neural Networks](https://mlsanity.com/paper/1311.1780/) (2013)
- [Would Mega-scale Datasets Further Enhance Spatiotemporal 3D CNNs?](https://mlsanity.com/paper/2004.04968/) (2020)

## Built on by (most cited first, 26 of 26)

- [AdaPool: Exponential Adaptive Pooling for Information-Retaining Downsampling](https://mlsanity.com/paper/2111.00772/) (2021)
- [Holomorphic Equilibrium Propagation Computes Exact Gradients Through Finite Size Oscillations](https://mlsanity.com/paper/2209.00530/) (2022)
- [Efficient Spatial-Temporal Information Fusion for LiDAR-Based 3D Moving Object Segmentation](https://mlsanity.com/paper/2207.02201/) (2022)
- [Temporal Lift Pooling for Continuous Sign Language Recognition](https://mlsanity.com/paper/2207.08734/) (2022)
- [Transformer-based Entity Typing in Knowledge Graphs](https://mlsanity.com/paper/2210.11151/) (2022)
- [Context-aware Entity Typing in Knowledge Graphs](https://mlsanity.com/paper/2109.07990/) (2021)
- [ICHPro: Intracerebral Hemorrhage Prognosis Classification Via Joint-attention Fusion-based 3d Cross-modal Network](https://mlsanity.com/paper/2402.11307/) (2024)
- [Improving equilibrium propagation without weight symmetry through Jacobian homeostasis](https://mlsanity.com/paper/2309.02214/) (2023)
- [NTIRE 2023 Quality Assessment of Video Enhancement Challenge](https://mlsanity.com/paper/2307.09729/) (2023)
- [EDEN: Deep Feature Distribution Pooling for Saimaa Ringed Seals Pattern Matching](https://mlsanity.com/paper/2105.13979/) (2021)
- [Content-Adaptive Downsampling in Convolutional Neural Networks](https://mlsanity.com/paper/2305.09504/) (2023)
- [FedRFQ: Prototype-Based Federated Learning with Reduced Redundancy, Minimal Failure, and Enhanced Quality](https://mlsanity.com/paper/2401.07558/) (2024)
- [Optimal Transport Guided Correlation Assignment for Multimodal Entity Linking](https://mlsanity.com/paper/2406.01934/) (2024)
- [HST-MRF: Heterogeneous Swin Transformer with Multi-Receptive Field for Medical Image Segmentation](https://mlsanity.com/paper/2304.04614/) (2023)
- [Forensic Histopathological Recognition via a Context-Aware MIL Network Powered by Self-Supervised Contrastive Learning](https://mlsanity.com/paper/2308.14030/) (2023)
- [A Comprehensive Review on Tree Detection Methods Using Point Cloud and Aerial Imagery from Unmanned Aerial Vehicles](https://mlsanity.com/paper/2309.16375/) (2023)
- [CrosswalkNet: An Optimized Deep Learning Framework for Pedestrian Crosswalk Detection in Aerial Images with High-Performance Computing](https://mlsanity.com/paper/2506.07885/) (2025)
- [DA-Font: Few-Shot Font Generation via Dual-Attention Hybrid Integration](https://mlsanity.com/paper/2509.16632/) (2025)
- [Representative Teacher Keys for Knowledge Distillation Model Compression Based on Attention Mechanism for Image Classification](https://mlsanity.com/paper/2206.12788/) (2022)
- [Dual Stream Computer-Generated Image Detection Network Based On Channel Joint And Softpool](https://mlsanity.com/paper/2207.03205/) (2022)
- [A Novel Dual-pooling Attention Module for UAV Vehicle Re-identification](https://mlsanity.com/paper/2306.14104/) (2023)
- [The Road to On-board Change Detection: A Lightweight Patch-Level Change Detection Network via Exploring the Potential of Pruning and Pooling](https://mlsanity.com/paper/2310.10166/) (2023)
- [Hybrid Pooling and Convolutional Network for Improving Accuracy and Training Convergence Speed in Object Detection](https://mlsanity.com/paper/2401.01134/) (2024)
- [MS-YOLO: A Multi-Scale Model for Accurate and Efficient Blood Cell Detection](https://mlsanity.com/paper/2506.03972/) (2025)
- [Redefining the Down-Sampling Scheme of U-Net for Precision Biomedical Image Segmentation](https://mlsanity.com/paper/2602.19412/) (2026)
- [Small Target Detection Based on Mask-Enhanced Attention Fusion of Visible and Infrared Remote Sensing Images](https://mlsanity.com/paper/2603.06925/) (2026)

## Abstract (arXiv)

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.