All Outputs (2)

ScribFormer: Transformer Makes CNN Work Better for Scribble-based Medical Image Segmentation (2024)
Journal Article
Li, Z., Zheng, Y., Shan, D., Yang, S., Li, Q., Wang, B., …Shen, D. (in press). ScribFormer: Transformer Makes CNN Work Better for Scribble-based Medical Image Segmentation. IEEE Transactions on Medical Imaging, https://doi.org/10.1109/tmi.2024.3363190

Most recent scribble-supervised segmentation methods commonly adopt a CNN framework with an encoder-decoder architecture. Despite its multiple benefits, this framework generally can only capture small-range feature dependency for the convolutional la... Read More about ScribFormer: Transformer Makes CNN Work Better for Scribble-based Medical Image Segmentation.

Using outlier elimination to assess learning-based correspondence matching methods (2024)
Journal Article
Ding, X., Luo, Y., Jie, B., Li, Q., & Cheng, Y. (2024). Using outlier elimination to assess learning-based correspondence matching methods. Information Sciences, 659, Article 120056. https://doi.org/10.1016/j.ins.2023.120056

Recently, deep learning (DL) technology has been widely used in correspondence matching. The learning-based models are usually trained on benign image pairs with partial overlaps. Since DL model is usually data-dependent, non-overlapping images may b... Read More about Using outlier elimination to assess learning-based correspondence matching methods.