Filters








11,528 Hits in 3.3 sec

Weakly supervised segmentation from extreme points [article]

Holger Roth, Ling Zhang, Dong Yang, Fausto Milletari, Ziyue Xu, Xiaosong Wang, Daguang Xu
2019 arXiv   pre-print
We use extreme points in each dimension of a 3D medical image to constrain an initial segmentation based on the random walker algorithm.  ...  Here, we propose to use minimal user interaction in the form of extreme point clicks in order to train a segmentation model that can, in turn, be used to speed up the annotation of medical images.  ...  Discussion & Conclusions We presented a method for weakly supervised 3D segmentation from extreme points.  ... 
arXiv:1910.01236v1 fatcat:joub4vndynfllbhkqma6fsdhim

Inter Extreme Points Geodesics for Weakly Supervised Segmentation [article]

Reuben Dorent, Samuel Joutard, Jonathan Shapey, Aaron Kujawa, Marc Modat, Sebastien Ourselin, Tom Vercauteren
2021 arXiv   pre-print
From the extreme points, 3D bounding boxes are extracted around objects of interest.  ...  We introduce InExtremIS, a weakly supervised 3D approach to train a deep image segmentation network using particularly weak train-time annotations: only 6 extreme clicks at the boundary of the objects  ...  In this work, we propose a novel weakly supervised approach to learn automatic image segmentation using extreme points as weak annotations during training, here a set of manual extreme points along each  ... 
arXiv:2107.00583v1 fatcat:ncozlnfitzdgbmd5r6t7qy635y

Going to Extremes: Weakly Supervised Medical Image Segmentation

Holger R. Roth, Dong Yang, Ziyue Xu, Xiaosong Wang, Daguang Xu
2021 Machine Learning and Knowledge Extraction  
An initial segmentation is generated based on the extreme points using the random walker algorithm.  ...  Here, we suggest using minimal user interaction in the form of extreme point clicks to train a segmentation model which, in effect, can be used to speed up medical image annotation.  ...  Effect of Extreme Point Noise during Inference Discussion We provided a method for weakly supervised 3D segmentation from extreme points.  ... 
doi:10.3390/make3020026 fatcat:vpy3rtl63rctjcv3sgtd7mn56u

Going to Extremes: Weakly Supervised Medical Image Segmentation [article]

Holger R Roth, Dong Yang, Ziyue Xu, Xiaosong Wang, Daguang Xu
2020 arXiv   pre-print
An initial segmentation is generated based on the extreme points utilizing the random walker algorithm.  ...  Here, we suggest using minimal user interaction in the form of extreme point clicks to train a segmentation model which, in effect, can be used to speed up medical image annotation.  ...  Discussion We provided a method for weakly-supervised 3D segmentation from extreme points.  ... 
arXiv:2009.11988v1 fatcat:msm35eefarharcaldcmjhr3wzm

BoxNet: Deep Learning Based Biomedical Image Segmentation Using Boxes Only Annotation [article]

Lin Yang, Yizhe Zhang, Zhuo Zhao, Hao Zheng, Peixian Liang, Michael T. C. Ying, Anil T. Ahuja, Danny Z. Chen
2018 arXiv   pre-print
To alleviate the burden of manual annotation, in this paper, we propose a new weakly supervised DL approach for biomedical image segmentation using boxes only annotation.  ...  performance over the best known state-of-the-art weakly supervised DL method and is able to achieve (1) nearly the same accuracy compared to fully supervised DL methods with far less annotation effort  ...  A set of weakly supervised annotation methods has been proposed for semantic segmentation in natural scene images.  ... 
arXiv:1806.00593v1 fatcat:hxyrku6qcfc4vljkjwikjp34du

Scribble2D5: Weakly-Supervised Volumetric Image Segmentation via Scribble Annotations [article]

Qiuhui Chen, Yi Hong
2022 arXiv   pre-print
Recently, weakly-supervised image segmentation using weak annotations like scribbles has gained great attention, since such annotations are much easier to obtain compared to time-consuming and label-intensive  ...  In this paper, we propose a scribble-based volumetric image segmentation, Scribble2D5, which tackles 3D anisotropic image segmentation and improves boundary prediction.  ...  (i.e., UNet PCE [17] and MAAG [18] ) and a weakly-supervised method using extreme points [8] .  ... 
arXiv:2205.06779v1 fatcat:o7y6ha62zbbiri3yfiy4mscfay

One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation [article]

Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu
2021 arXiv   pre-print
Experimental results on both ScanNet-v2 and S3DIS show that our self-training approach, with extremely-sparse annotations, outperforms all existing weakly supervised methods for 3D semantic segmentation  ...  Point cloud semantic segmentation often requires largescale annotated training data, but clearly, point-wise labels are too tedious to prepare.  ...  Weakly Supervised 3D Semantic Segmentation Compared with fully supervised 3D semantic segmentation, weakly supervised 3D semantic segmentation is relatively under-explored.  ... 
arXiv:2104.02246v4 fatcat:ucniyqpo6zgi5k4ned7my5nafq

Weakly Supervised Multiclass Video Segmentation

Xiao Liu, Dacheng Tao, Mingli Song, Ying Ruan, Chun Chen, Jiajun Bu
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
In this paper, we present a novel nearest neighbor-based label transfer scheme for weakly supervised video segmentation.  ...  Whereas previous weakly supervised video segmentation methods have been limited to the two-class case, our proposed scheme focuses on more challenging multiclass video segmentation, which finds a semantically  ...  Weakly Supervised Multiclass Video Segmentation Weakly labeled videos are first segmented into spatiotemporal supervoxels, which are represented as high dimensional points in the feature space, and weakly  ... 
doi:10.1109/cvpr.2014.15 dblp:conf/cvpr/LiuTSRCB14 fatcat:wdbnf4p2vjh7bpy23j25ahhb5y

One-shot Weakly-Supervised Segmentation in Medical Images [article]

Wenhui Lei, Qi Su, Ran Gu, Na Wang, Xinglong Liu, Guotai Wang, Xiaofan Zhang, Shaoting Zhang
2021 arXiv   pre-print
One-shot segmentation and weakly-supervised learning are promising research directions that lower labeling effort by learning a new class from only one annotated image and utilizing coarse labels instead  ...  Hence, we present an innovative framework for 3D medical image segmentation with one-shot and weakly-supervised settings.  ...  Weakly Supervised Segmentation Compared with fully supervised segmentation task which needs time-consuming pixel-wise annotations, WSS requires more flexible annotations like scribbles [32] , bounding  ... 
arXiv:2111.10773v1 fatcat:bzryc4hkqnabbifopr3icimdku

PCAMs: Weakly Supervised Semantic Segmentation Using Point Supervision [article]

R. Austin McEver, B.S. Manjunath
2020 arXiv   pre-print
This paper presents a novel procedure for producing semantic segmentation from images given some point level annotations.  ...  Several methods have shown that we can learn semantic segmentation from less expensive image-level labels, but the effectiveness of point level labels, a healthy compromise between all pixels labelled  ...  Conclusion While numerous methods for using image-level supervision for weakly supervised semantic segmentation have been introduced recently, using point supervision has been largely unexplored.  ... 
arXiv:2007.05615v1 fatcat:qa3locldtjcajgicvnt4ttnfii

GECNN for Weakly Supervised Semantic Segmentation of 3D Point Clouds

Zifen HE, Shouye ZHU, Ying HUANG, Yinhui ZHANG
2021 IEICE transactions on information and systems  
This paper presents a novel method for weakly supervised semantic segmentation of 3D point clouds using a novel graph and edge convolutional neural network (GECNN) towards 1% and 10% point cloud with labels  ...  It inherently handles the challenges of segmenting sparse 3D point clouds with limited annotations in a large scale point cloud space.  ...  [4] propose to employ conditional random field for weakly supervised segmentation of urban scenes from 3D LiDAR point clouds.  ... 
doi:10.1587/transinf.2021edp7134 fatcat:iy6ctacj4rddvgfbltcuvybwb4

Medical image segmentation with imperfect 3D bounding boxes [article]

Ekaterina Redekop, Alexey Chernyavskiy
2021 arXiv   pre-print
When the tightness is improved by our solution, the results of the weakly-supervised segmentation become much closer to those of the fully-supervised one.  ...  The effectiveness of our solution is demonstrated by evaluating a known weakly-supervised segmentation approach with and without the proposed bounding box correction algorithm.  ...  It is straightforward to produce a 3D bounding box of the object by finding its extreme points in the three coordinate axes.  ... 
arXiv:2108.03300v1 fatcat:an4te3vgnncbdkf3xul336zqom

Weakly Supervised 3D Point Cloud Segmentation via Multi-Prototype Learning [article]

Yongyi Su, Xun Xu, Kui Jia
2022 arXiv   pre-print
Addressing the annotation challenge in 3D Point Cloud segmentation has inspired research into weakly supervised learning.  ...  Experiments on weakly supervised 3D point cloud segmentation tasks validate the efficacy of proposed method in particular at low-label regime.  ...  Architecture Overview To formally define the weakly-supervised 3D point segmentation task, we follow the settings proposed in [1] .  ... 
arXiv:2205.03137v1 fatcat:bdu7ytk5pjaqnb62dzhtznnlom

WAILS: Watershed Algorithm with Image-level Supervision for Weakly Supervised Semantic Segmentation

Hongming Zhou, Kang Song, Xianglei Zhang, Wenyong Gui, Qiusuo Qian
2019 IEEE Access  
INDEX TERMS Semantic segmentation, weakly supervised, watershed algorithm.  ...  First, the image is coarsely segmented through a weakly supervised network at the image level.  ...  Some classic weakly supervised image semantic segmentation label methods include point, scribble, and bounding box. For example, Bearman et al. [10] and Kwak et al.  ... 
doi:10.1109/access.2019.2908216 fatcat:tj55fscpqbfmvbpiftdrhkrkme

Point-supervised Segmentation of Microscopy Images and Volumes via Objectness Regularization [article]

Shijie Li, Neel Dey, Katharina Bermond, Leon von der Emde, Christine A. Curcio, Thomas Ach, Guido Gerig
2021 arXiv   pre-print
This work enables the training of semantic segmentation networks on images with only a single point for training per instance, an extreme case of weak supervision which drastically reduces the burden of  ...  We achieve competitive results against the state-of-the-art in point-supervised semantic segmentation on challenging datasets in digital pathology.  ...  which enables weakly supervised training.  ... 
arXiv:2103.05617v2 fatcat:h3phricp6jfqrcynsqxf2e4vui
« Previous Showing results 1 — 15 out of 11,528 results