Parallel matters : Efficient polyp segmentation with parallel structured feature augmentation modules
(2023) In IET Image Processing 17(8). p.2503-2515- Abstract
The large variations of polyp sizes and shapes and the close resemblances of polyps to their surroundings call for features with long-range information in rich scales and strong discrimination. This article proposes two parallel structured modules for building those features. One is the Transformer Inception module (TI) which applies Transformers with different reception fields in parallel to input features and thus enriches them with more long-range information in more scales. The other is the Local-Detail Augmentation module (LDA) which applies the spatial and channel attentions in parallel to each block and thus locally augments the features from two complementary dimensions for more object details. Integrating TI and LDA, a new... (More)
The large variations of polyp sizes and shapes and the close resemblances of polyps to their surroundings call for features with long-range information in rich scales and strong discrimination. This article proposes two parallel structured modules for building those features. One is the Transformer Inception module (TI) which applies Transformers with different reception fields in parallel to input features and thus enriches them with more long-range information in more scales. The other is the Local-Detail Augmentation module (LDA) which applies the spatial and channel attentions in parallel to each block and thus locally augments the features from two complementary dimensions for more object details. Integrating TI and LDA, a new Transformer encoder based framework, Parallel-Enhanced Network (PENet), is proposed, where LDA is specifically adopted twice in a coarse-to-fine way for accurate prediction. PENet is efficient in segmenting polyps with different sizes and shapes without the interference from the background tissues. Experimental comparisons with state-of-the-arts methods show its merits.
(Less)
- author
- Guo, Qingqing ; Fang, Xianyong ; Wang, Kaibing ; Shi, Yuqing ; Wang, Linbo ; Zhang, Enming LU and Liu, Zhengyi
- organization
- publishing date
- 2023
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- biomedical imaging, computer vision, image segmentation
- in
- IET Image Processing
- volume
- 17
- issue
- 8
- pages
- 13 pages
- publisher
- John Wiley & Sons Inc.
- external identifiers
-
- scopus:85153480316
- ISSN
- 1751-9659
- DOI
- 10.1049/ipr2.12813
- language
- English
- LU publication?
- yes
- id
- 50b46a7c-3478-471d-8efe-6573ea0109b2
- date added to LUP
- 2023-07-14 13:53:31
- date last changed
- 2023-11-08 07:39:53
@article{50b46a7c-3478-471d-8efe-6573ea0109b2, abstract = {{<p>The large variations of polyp sizes and shapes and the close resemblances of polyps to their surroundings call for features with long-range information in rich scales and strong discrimination. This article proposes two parallel structured modules for building those features. One is the Transformer Inception module (TI) which applies Transformers with different reception fields in parallel to input features and thus enriches them with more long-range information in more scales. The other is the Local-Detail Augmentation module (LDA) which applies the spatial and channel attentions in parallel to each block and thus locally augments the features from two complementary dimensions for more object details. Integrating TI and LDA, a new Transformer encoder based framework, Parallel-Enhanced Network (PENet), is proposed, where LDA is specifically adopted twice in a coarse-to-fine way for accurate prediction. PENet is efficient in segmenting polyps with different sizes and shapes without the interference from the background tissues. Experimental comparisons with state-of-the-arts methods show its merits.</p>}}, author = {{Guo, Qingqing and Fang, Xianyong and Wang, Kaibing and Shi, Yuqing and Wang, Linbo and Zhang, Enming and Liu, Zhengyi}}, issn = {{1751-9659}}, keywords = {{biomedical imaging; computer vision; image segmentation}}, language = {{eng}}, number = {{8}}, pages = {{2503--2515}}, publisher = {{John Wiley & Sons Inc.}}, series = {{IET Image Processing}}, title = {{Parallel matters : Efficient polyp segmentation with parallel structured feature augmentation modules}}, url = {{http://dx.doi.org/10.1049/ipr2.12813}}, doi = {{10.1049/ipr2.12813}}, volume = {{17}}, year = {{2023}}, }