search

SwinTransformer / Swin-Transformer-Object-Detection

This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows" on Object Detection and Instance Segmentation.
https://arxiv.org/abs/2103.14030
Apache License 2.0
1.79k stars 378 forks source link

Model usage in an FPN-architecture #7

Closed kzhang0718 closed 3 years ago

kzhang0718 commented 3 years ago

Hi,

Just looking for some advice on how to use the current implementation of Swin Transformer in an FPN-based detector model. Does the current implementation work out of the box, or some modifications to the model must be done?

Thanks.

impiga commented 3 years ago

Hi, @kzhang0718. Our current implementation is suitable for FPN architecture.

To use Swin Transformer in a FPN-based model, we need to modify the config file accordingly. The following is an example: https://github.com/SwinTransformer/Swin-Transformer-Object-Detection/blob/02baa301774b455fe8e57ef9dc0fc32ef2ba2bb6/configs/_base_/models/cascade_mask_rcnn_swin_fpn.py#L5-L25