Semantic Segmentation Models
Criss-Cross Network
Introduced by Huang et al. in CCNet: Criss-Cross Attention for Semantic SegmentationCriss-Cross Network (CCNet) aims to obtain full-image contextual information in an effective and efficient way. Concretely, for each pixel, a novel criss-cross attention module harvests the contextual information of all the pixels on its criss-cross path. By taking a further recurrent operation, each pixel can finally capture the full-image dependencies. CCNet is with the following merits: 1) GPU memory friendly. Compared with the non-local block, the proposed recurrent criss-cross attention module requires 11× less GPU memory usage. 2) High computational efficiency. The recurrent criss-cross attention significantly reduces FLOPs by about 85% of the non-local block. 3) The state-of-the-art performance.
Source: CCNet: Criss-Cross Attention for Semantic Segmentation
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Scene Understanding | 1 | 7.69% |
| Deblurring | 1 | 7.69% |
| Image Dehazing | 1 | 7.69% |
| Image Restoration | 1 | 7.69% |
| Common Sense Reasoning | 1 | 7.69% |
| Retrieval | 1 | 7.69% |
| Computational Efficiency | 1 | 7.69% |
| Human Parsing | 1 | 7.69% |
| Instance Segmentation | 1 | 7.69% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
