Instance Segmentation Models
Hybrid Task Cascade
Introduced by Chen et al. in Hybrid Task Cascade for Instance SegmentationHybrid Task Cascade, or HTC, is a framework for cascading in instance segmentation. It differs from Cascade Mask R-CNN in two important aspects: (1) instead of performing cascaded refinement on the two tasks of detection and segmentation separately, it interweaves them for a joint multi-stage processing; (2) it adopts a fully convolutional branch to provide spatial context, which can help distinguishing hard foreground from cluttered background.
Source: Hybrid Task Cascade for Instance Segmentation
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Text Classification | 17 | 16.83% |
| Classification | 9 | 8.91% |
| Multi-Label Classification | 8 | 7.92% |
| MUlTI-LABEL-ClASSIFICATION | 7 | 6.93% |
| Instance Segmentation | 7 | 6.93% |
| Semantic Segmentation | 6 | 5.94% |
| Language Modeling | 4 | 3.96% |
| Language Modelling | 4 | 3.96% |
| Object | 4 | 3.96% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
1x1 Convolution
|
Convolutions | |
Convolution
|
Convolutions | |
FPN
|
Feature Extractors | |
RoIAlign
|
RoI Feature Extractors | |
RPN
|
Region Proposal |
