简体   繁体   English

用于 3D 分割的深度学习

[英]deep learning for 3d segmentation

I have 20 heart images taken at the same location and there are 50 of those images.我在同一位置拍摄了 20 张心脏图像,其中有 50 张图像。 So, 20 images are one input and I have 50 inputs.因此,20 张图像是一个输入,而我有 50 个输入。 Because heart is moving, all images are different.因为心在动,所有的图像都不一样。 I want to make those images as input and then segmented binary mask images as output using deep learning.我想将这些图像作为输入,然后使用深度学习将二进制掩码图像分割为输出。

I treat my inputs as 3D and found a lot of 3D CNN or FCN online.我将我的输入视为 3D,并在网上找到了很多 3D CNN 或 FCN。 But those are the case of x,y,z 3D.但这些是 x,y,z 3D 的情况。 My case is x, y, t.我的情况是x,y,t。

Any suggestion?有什么建议吗?

This problem is similar to video object segmentation , where an object needs to be semantically segmented across several frames in a video by utilizing temporal information across frames.这个问题类似于视频对象分割,其中需要利用跨帧的时间信息对视频中的多个帧进行语义分割。 In your case, each set of input image can be thought of as a video with 20 frames with dataset containing 50 videos and you can use supervised, semi-supervised or unsupervised video object segmentation models (depending on the availability/cost of obtaining ground truth masks).在您的情况下,每组输入图像都可以被认为是一个20 帧的视频,数据集包含50 个视频,您可以使用有监督、半监督或无监督的视频对象分割模型(取决于获取地面实况的可用性/成本)面具)。 This formulation solves the task of segmenting input image stack in (x,y,t) order with FCNs.该公式解决了使用 FCN 按(x,y,t)顺序分割输入图像堆栈的任务。

Below are references I found for research work related to video object segmentation:以下是我在与视频对象分割相关的研究工作中找到的参考资料:

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM