Reinventing 2D Convolutions for 3D Images

Reinventing 2D Convolutions for 3D Images

There have been considerable debates over 2D and 3D representation learning on 3D medical images. 2D approaches could benefit from large-scale 2D pretraining, whereas they are generally weak in capturing large 3D contexts. 3D approaches are natively strong in 3D contexts, however few publicly availa...

Full description

Bibliographic Details
Main Authors: He, Y. (Author), Huang, X. (Author), Ni, B. (Author), Xu, G. (Author), Xu, J. (Author), Yang, C. (Author), Yang, J. (Author)
Format: Article
Language:English
Published: Institute of Electrical and Electronics Engineers Inc. 2021
Subjects:
2-D convolution
2D-to-3D transfer learning
3D medical image
3D medical images
3d representations
ACS convolutions
algorithm
Algorithms
article
Convolution
deep learning
Detection tasks
human
Humans
Imaging, Three-Dimensional
Large dataset
Medical dataset
Medical imaging
Plug and play
Pre-training
Three views
three-dimensional imaging
transfer of learning
videorecording
Online Access:View Fulltext in Publisher

Internet

View Fulltext in Publisher

Similar Items