nnFormer: Interleaved Transformer for Volumetric Segmentation
By Hong-Yu Zhou, Jiansen Guo, Yinghao Zhang, Lequan Yu, Liansheng Wang, Yizhou Yu. arXiv.org, 2021
476 citations recorded by the source corpus.
View Paper: https://www.semanticscholar.org/paper/31444a27bb09e64cf252be9fa2349a5178307012
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Salman Hameed Khan
- Fahad Shahbaz Khan
- Ming-Hsuan Yang
- Syed Waqas Zamir
- Munawar Hayat
- Luc Van Gool
- Martin Danelljan
- Wenguan Wang
Citations in the corpus
- A Comprehensive Survey on Applications of Transformers for Deep Learning Tasks (2023, methodology): “Additionally, nnFormer employs multiple convolution layers with small kernels in the encoder instead of large convolution kernels as in other visual transformers (Zhou et al., 2021a).”
- AMOS: A Large-Scale Abdominal Multi-Organ Benchmark for Versatile Medical Image Segmentation (2022, methodology): “In this group of work [28, 6, 7, 24], the features patch or image patch are treated as tokens, which are used to conduct efficient non-local context modeling among arbitrary positions, making them achieve SOTA performance on the most popular dataset [12].”
- Transformers in Medical Imaging: A Survey (2022, background): “[144] propose nnFormer, an interleave encoder-decoder based architecture, where convolution layer encodes precise spatial information and Transformer layer encodes global context as shown in Fig.”
- CLIP-Driven Universal Model for Organ Segmentation and Tumor Detection (2023, background): “Recently, transformer based models [7, 83, 21, 64, 6] are emerged, which can capture the global relationship between whole volume.”
- SegMamba: Long-range Sequential Modeling Mamba For 3D Medical Image Segmentation (2024, methodology): “We run 1000 epochs for all datasets and adopt the following data augmentations: additive brightness, gamma, rotation, scaling, mirror, and elastic deformation.”
- nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentation (2024, methodology): “Further, the practice of reporting selective results only for specific classes from datasets without justifiable reasons further compromises result integrity [41,11,9,36,39].”
- UNETR++: Delving Into Efficient and Accurate 3D Medical Image Segmentation (2022, methodology): “Consistent with [36], we split the data into 70, 10 and 20 train, validation and test samples.”
- Semi-Supervised Medical Image Segmentation via Cross Teaching between CNN and Transformer (2021, background): “…of transformers, several works attempt to use transformers to replace or combine CNNs for better medical image segmentation results, such as TransUNet (Chen et al., 2021a), Swin-UNet (Cao et al., 2021), CoTr (Xie et al., 2021), UN-ETR (Hatamizadeh et al., 2021), nnFormer (Zhou et al., 2021), etc.”
- TransMorph: Transformer for unsupervised medical image registration (2021, unknown): “• nnFormer 15 (Zhou et al., 2021): Because nnFormer was also developed on the basis of Swin Transformer, we applied the same Transformer hyperparameter values as in TransMorph to make a fair comparison.”