无法在COCO注释器中创建数据集并加载图像 [英] Can't create datasets and load images in COCO annotator
问题描述
我正在尝试使用 https://github来对带有COCO关键点的图像进行姿势估计.com/jsbroks/coco-annotator .如安装"部分所述,我克隆了存储库.我安装了Docker和Docker-compose.之后,我使用 $ docker-compose up
启动了容器,并且该容器正在运行.
I'm trying to annotate images with COCO key points for pose estimation using https://github.com/jsbroks/coco-annotator. As described in the Installation section I cloned the repo. I installed Docker and Docker-compose. Following this I started the container with $ docker-compose up
and it is running.
我现在在网站 https://annotator.justinbrooks.ca/上,我创建了一个用户和数据集,但它们不会出现在repo datasets/
文件夹中.我尝试手动创建它们并在其中加载图像,但它们未出现在网站图形界面上.
I am now on the website https://annotator.justinbrooks.ca/, I created one user and datasets but they do not appear in the repo datasets/
folder. I tried to create them manually and to load images in them but they do not appear on the website graphic interface.
我试图扫描,重新加载网页,创建其他数据集,但是它不起作用.该容器似乎可以正常工作,它可以检测到何时将图像放入 datasets/
文件夹中,但是会引发一些错误.
I tried to scan, reload the webpage, create other datasets but it does not work. The container seems to work properly, it detects when I put an image in the datasets/
folder but it throws some errors.
这是最后几行(我可以发布整个日志):
Here is the last lines (I can post the whole log):
annotator_webclient | [File Watcher] File /datasets/haricot.jpg for created
annotator_webclient | [File Watcher] Adding new file to database: /datasets/image
annotator_message_q | 2019-05-16 13:01:08.841 [error] <0.461.0> closing AMQP connection <0.461.0> (172.18.0.4:42614 -> 172.18.0.2:5672):
annotator_message_q | missed heartbeats from client, timeout: 60s
我是否缺少基本知识或存在错误?
Am I missing something fundamental or there is a bug?
我在macOS上使用Safari,也曾在Ubuntu 18上尝试过Firefox.我没有使用代理,但是某些端口未打开或类似的东西.
I'm using Safari on macOS and I also tried Firefox on Ubuntu 18. I'm not behind a proxy, but maybe some ports are not open or something like this.
推荐答案
此处创建COCO注释器.
Creator of COCO Annotator here.
我认为您缺少一个基本概念.该演示在VM上运行,与您的本地实例无关.
I think you are missing a fundamental concept. The demo runs on the VM and has nothing to do with your local instance.
运行了docker-compose之后,您可以通过输入 http://localhost:5000/来访问本地实例.a>.
Once you have the docker-compose running you can access your instance local by typing http://localhost:5000/ .
这篇关于无法在COCO注释器中创建数据集并加载图像的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!