Better interactive image segmentation for training by using edge information
Better interactive image segmentation for training by using edge information
EdgeFlow: Achieving Practical Interactive Segmentation with Edge-Guided Flow
arXiv paper abstract https://arxiv.org/abs/2109.09406v1
arXiv PDF paper https://arxiv.org/pdf/2109.09406v1.pdf
GitHub https://github.com/PaddlePaddle/PaddleSeg
High-quality training data play a key role in image segmentation tasks.
Usually, pixel-level annotations are expensive, laborious and time-consuming for the large volume of training data.
… interactive segmentation methods have been proposed … However, their performance does not meet the requirements of practical segmentation tasks in terms of speed and accuracy.
… propose EdgeFlow, a novel architecture that fully utilizes interactive information of user clicks with edge-guided flow.
… achieves state-of-the-art performance without any post-processing or iterative optimization scheme. …
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
