Simplified unsupervised image translation for semantic segmentation adaptation

Li, Rui; Cao, Wenming; Jiao, Qianfen; Wu, Si; Wong, Hau-San

doi:10.1016/j.patcog.2020.107343

Cited by 53 publications

(16 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [25], a improved Unsupervised Image Translation (SUIT) method is described for semantic segmentation. They adopt adversarial training for superior image generation, and design a novel semantic-content loss to enhance visual appearance preservation.…”

Section: Potential Methods For Casamentioning

confidence: 99%

A Survey of Semen Quality Evaluation in Microscopic Videos Using Computer Assisted Sperm Analysis

Zhao¹,

Ma²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

The Computer Assisted Sperm Analysis (CASA) plays a crucial role in male reproductive health diagnosis and Infertility treatment. With the development of the computer industry in recent years, a great of accurate algorithms are proposed. With the assistance of those novel algorithms, it is possible for CASA to achieve a faster and higher quality result. Since image processing is the technical basis of CASA, including pre-processing,feature extraction, target detection and tracking, these methods are important technical steps in dealing with CASA. The various works related to Computer Assisted Sperm Analysis methods in the last 30 years (since 1988) are comprehensively introduced and analysed in this survey. To facilitate understanding, the methods involved are analysed in the sequence of general steps in sperm analysis. In other words, the methods related to sperm detection (localization) are first analysed, and then the methods of sperm tracking are analysed. Beside this, we analyse and prospect the present situation and future of CASA. According to our work, the feasible for applying in sperm microscopic video of methods mentioned in this review is explained. Moreover, existing challenges of object detection and tracking in microscope video are potential to be solved inspired by this survey.

show abstract

Section: Potential Methods For Casamentioning

confidence: 99%

A Survey of Semen Quality Evaluation in Microscopic Videos Using Computer Assisted Sperm Analysis

Zhao¹,

Ma²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Typically, generative adversarial networks (GAN) ( 12 ) are utilized to learn a distribution over the target images, conditioned by the input image ( 5 ). Such approaches have been applied to a wide range of topics, for example semantic image synthesis ( 13 ), image segmentation ( 14 ), style transfer ( 15 ), or image inpainting ( 16 ). However, most of these approaches are used in settings with a moderate domain gap while the general objects in the input image remain unchanged in their shape.…”

Section: Related Workmentioning

confidence: 99%

Aortic Valve Leaflet Shape Synthesis With Geometric Prior From Surrounding Tissue

Hagenah

Scharfschwerdt

Ernst

2022

Front. Cardiovasc. Med.

View full text Add to dashboard Cite

Even though the field of medical imaging advances, there are structures in the human body that are barely assessible with classical image acquisition modalities. One example are the three leaflets of the aortic valve due to their thin structure and high movement. However, with an increasing accuracy of biomechanical simulation, for example of the heart function, and extense computing capabilities available, concise knowledge of the individual morphology of these structures could have a high impact on personalized therapy and intervention planning as well as on clinical research. Thus, there is a high demand to estimate the individual shape of inassessible structures given only information on the geometry of the surrounding tissue. This leads to a domain adaptation problem, where the domain gap could be very large while typically only small datasets are available. Hence, classical approaches for domain adaptation are not capable of providing sufficient predictions. In this work, we present a new framework for bridging this domain gap in the scope of estimating anatomical shapes based on the surrounding tissue's morphology. Thus, we propose deep representation learning to not map from one image to another but to predict a latent shape representation. We formalize this framework and present two different approaches to solve the given problem. Furthermore, we perform a proof-of-concept study for estimating the individual shape of the aortic valve leaflets based on a volumetric ultrasound image of the aortic root. Therefore, we collect an ex-vivo porcine data set consisting of both, ultrasound volume images as well as high-resolution leaflet images, evaluate both approaches on it and perform an analysis of the model's hyperparameters. Our results show that using deep representation learning and domain mapping between the identified latent spaces, a robust prediction of the unknown leaflet shape only based on surrounding tissue information is possible, even in limited data scenarios. The concept can be applied to a wide range of modeling tasks, not only in the scope of heart modeling but also for all kinds of inassessible structures within the human body.

show abstract

“…Compared with traditional image segmentation methods, semantic segmentation methods based on DL automatically learn features from data instead of manually designed features. The fully convolutional network (FCN) proposed in [23] removes the fully connected layer on the basis of the image classification network visual geometry group (VGG) network, adds multi-level upsampling to restore the resolution and achieves end-to-end Semantic segmentation; Li et al [24] proposed a simplified unsupervised image translation (SUIT) model for domain adaptation of semantic segmentation, avoiding manual dense pixel-level labelling and achieved good results; Many algorithms downsample the input image multiple times in order to increase the receptive field, reduce the feature dimension and reduce the amount of calculation. However, in this process, the constraint of the loss function on the features is getting lower and lower, resulting in low dispersion of low-level features.…”

Section: Related Workmentioning

confidence: 99%

Image semantic segmentation method based on GAN network and ENet model

Li¹

2021

J. eng.

View full text Add to dashboard Cite

Currently, image semantic segmentation has problems such as low accuracy and long running time. This paper proposes an image semantic segmentation method based on generative adversarial network and ENet model combined with deep neural network. This method first improves the network model of generative adversarial network. Ensure the high resolution of the generated image and achieve high similarity with the real image. While ensuring the high accuracy of image semantic segmentation, it effectively improves the real-time performance of network processing. The proposed method is verified based on public data sets. The experimental results show that the segmentation accuracy of this method can reach more than 93%, and the simulation running time is less than 0.171 s, which shows good high accuracy and high real-time performance. A feasible strategy is proposed for the further productization of semantic segmentation.

show abstract

Simplified unsupervised image translation for semantic segmentation adaptation

Cited by 53 publications

References 7 publications

A Survey of Semen Quality Evaluation in Microscopic Videos Using Computer Assisted Sperm Analysis

A Survey of Semen Quality Evaluation in Microscopic Videos Using Computer Assisted Sperm Analysis

Aortic Valve Leaflet Shape Synthesis With Geometric Prior From Surrounding Tissue

Image semantic segmentation method based on GAN network and ENet model

Contact Info

Product

Resources

About