Max Pooling is a pooling operation that calculates the maximum value for patches of a feature map, and uses it to create a downsampled (pooled) feature map. It is usually used after a convolutional layer. It adds a small amount of translation invariance - meaning translating the image by a small amount does not significantly affect the values of most pooled outputs.
Image Source: here
Paper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Semantic Segmentation | 71 | 9.57% |
Image Segmentation | 38 | 5.12% |
Denoising | 31 | 4.18% |
Image Generation | 30 | 4.04% |
Image Classification | 28 | 3.77% |
Self-Supervised Learning | 21 | 2.83% |
Medical Image Segmentation | 18 | 2.43% |
Object Detection | 18 | 2.43% |
Classification | 17 | 2.29% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |