Anti-Alias Downsampling (AA) aims to improve the shift-equivariance of deep networks. Max-pooling is inherently composed of two operations. The first operation is to densely evaluate the max operator and second operation is naive subsampling. AA is proposed as a low-pass filter between them to achieve practical anti-aliasing in any existing strided layer such as strided convolution. The smoothing factor can be adjusted by changing the blur kernel filter size, where a larger filter size results in increased blur.
Source: Making Convolutional Networks Shift-Invariant AgainPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Image Classification | 3 | 18.75% |
Fine-Grained Image Classification | 2 | 12.50% |
General Classification | 2 | 12.50% |
Speech Synthesis | 1 | 6.25% |
Text-To-Speech Synthesis | 1 | 6.25% |
Multi-Label Classification | 1 | 6.25% |
Object Detection | 1 | 6.25% |
Fine-Grained Visual Recognition | 1 | 6.25% |
Image Retrieval | 1 | 6.25% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |