Image Model Blocks
Local Patch Interaction
Introduced by El-Nouby et al. in XCiT: Cross-Covariance Image TransformersLocal Patch Interaction, or LPI, is a module used for the XCiT layer to enable explicit communication across patches. LPI consists of two depth-wise 3×3 convolutional layers with Batch Normalization and GELU non-linearity in between. Due to its depth-wise structure, the LPI block has a negligible overhead in terms of parameters, as well as a limited overhead in terms of throughput and memory usage during inference.
Source: XCiT: Cross-Covariance Image Transformers
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Image Classification | 3 | 30.00% |
| Quantization | 1 | 10.00% |
| Decoder | 1 | 10.00% |
| Pose Estimation | 1 | 10.00% |
| Instance Segmentation | 1 | 10.00% |
| Object Detection | 1 | 10.00% |
| Self-Supervised Image Classification | 1 | 10.00% |
| Semantic Segmentation | 1 | 10.00% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
Batch Normalization
|
Normalization | |
Depthwise Convolution
|
Convolutions | |
GELU
|
Activation Functions |
