Image augmentation using DETR transformer to select a patch with DeMix
Image augmentation using DETR transformer to select a patch with DeMix
Use the Detection Transformer as a Data Augmenter
arXiv paper abstract https://arxiv.org/abs/2304.04554
arXiv PDF paper https://arxiv.org/pdf/2304.04554.pdf
GitHub https://github.com/ZJLAB-AMMI/DeMix
Detection Transformer (DETR) is a Transformer architecture based object detection model … that it can also be used as a data augmenter … term … approach … DeMix …
DeMix builds on CutMix … augmentation … improves model performance by cutting and pasting a patch from one image onto another, yielding a new image.
… label … is … weighted average of … labels, where the weight is proportional to the area of the patches.
CutMix selects a random patch to be cut … In contrast, DeMix elaborately selects a semantically rich patch, located by a pre-trained DETR.
The label of the new image is specified in the same way as in CutMix.
… results on … image classification demonstrate that DeMix significantly outperforms prior art data augmentation methods including CutMix.
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b
