VoVNetV2 is a convolutional neural network that improves upon VoVNet with two effective strategies: (1) residual connection for alleviating the optimization problem of larger VoVNets and (2) effective Squeeze-Excitation (eSE) dealing with the channel information loss problem of the original squeeze-and-excitation module.
Source: CenterMask : Real-Time Anchor-Free Instance SegmentationPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Instance Segmentation | 1 | 14.29% |
Object Detection | 1 | 14.29% |
Panoptic Segmentation | 1 | 14.29% |
Real-time Instance Segmentation | 1 | 14.29% |
Real-Time Object Detection | 1 | 14.29% |
Semantic Segmentation | 1 | 14.29% |
Semi-Supervised Instance Segmentation | 1 | 14.29% |
Component | Type |
|
---|---|---|
Convolution
|
Convolutions | |
Max Pooling
|
Pooling Operations | |
OSA (identity mapping + eSE)
|
Skip Connection Blocks |