Mask RCNN detecting object but mask is inaccurate
$begingroup$
I am trying to detect the inner region of a object. Currently I am using the mask rcnn implementation provided by tensorflow in the models zoo.
The object is similar to a hula hoop that is square in shape. 
I tried the inception_V2 backbone and have about 3000 images from different angles and lighting conditions.
After about 20,000 steps the model seems to converege to ~0.2 (seems reasonable based on other examples of mask rcnn on the internet) and stay there for as long as I run it. I have tried upto 200,000 steps.
The problem I am having is that while the bounding box is correctly drawn, the mask is inaccurate. Instead of the ground truth mask which has well defined edges, I get wavy edges and overshoots/undershoots. (
My goal is to find the 4 corner points of the inner area and am unable to. Any suggestions? Is there a way to post process these masks?
Is it also possible to find the inner edges maybe from the bounding box? I tried classical methods but could not manage to get the inner area.
machine-learning image-recognition faster-rcnn
asked 6 mins ago
John DoeJohn Doe
1
New contributor
John Doe is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
$endgroup$
add a comment |
$begingroup$
I am trying to detect the inner region of a object. Currently I am using the mask rcnn implementation provided by tensorflow in the models zoo.
The object is similar to a hula hoop that is square in shape.
I tried the inception_V2 backbone and have about 3000 images from different angles and lighting conditions.
After about 20,000 steps the model seems to converege to ~0.2 (seems reasonable based on other examples of mask rcnn on the internet) and stay there for as long as I run it. I have tried upto 200,000 steps.
The problem I am having is that while the bounding box is correctly drawn, the mask is inaccurate. Instead of the ground truth mask which has well defined edges, I get wavy edges and overshoots/undershoots. (
My goal is to find the 4 corner points of the inner area and am unable to. Any suggestions? Is there a way to post process these masks?
Is it also possible to find the inner edges maybe from the bounding box? I tried classical methods but could not manage to get the inner area.
machine-learning image-recognition faster-rcnn
asked 6 mins ago
John DoeJohn Doe
1
New contributor
John Doe is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
$endgroup$
add a comment |
$begingroup$
I am trying to detect the inner region of a object. Currently I am using the mask rcnn implementation provided by tensorflow in the models zoo.
The object is similar to a hula hoop that is square in shape.
I tried the inception_V2 backbone and have about 3000 images from different angles and lighting conditions.
After about 20,000 steps the model seems to converege to ~0.2 (seems reasonable based on other examples of mask rcnn on the internet) and stay there for as long as I run it. I have tried upto 200,000 steps.
The problem I am having is that while the bounding box is correctly drawn, the mask is inaccurate. Instead of the ground truth mask which has well defined edges, I get wavy edges and overshoots/undershoots. (
My goal is to find the 4 corner points of the inner area and am unable to. Any suggestions? Is there a way to post process these masks?
Is it also possible to find the inner edges maybe from the bounding box? I tried classical methods but could not manage to get the inner area.
machine-learning image-recognition faster-rcnn
asked 6 mins ago
John DoeJohn Doe
1
New contributor
John Doe is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
$endgroup$
I am trying to detect the inner region of a object. Currently I am using the mask rcnn implementation provided by tensorflow in the models zoo.
The object is similar to a hula hoop that is square in shape.
I tried the inception_V2 backbone and have about 3000 images from different angles and lighting conditions.
After about 20,000 steps the model seems to converege to ~0.2 (seems reasonable based on other examples of mask rcnn on the internet) and stay there for as long as I run it. I have tried upto 200,000 steps.
The problem I am having is that while the bounding box is correctly drawn, the mask is inaccurate. Instead of the ground truth mask which has well defined edges, I get wavy edges and overshoots/undershoots. (
My goal is to find the 4 corner points of the inner area and am unable to. Any suggestions? Is there a way to post process these masks?
Is it also possible to find the inner edges maybe from the bounding box? I tried classical methods but could not manage to get the inner area.
machine-learning image-recognition faster-rcnn
machine-learning image-recognition faster-rcnn
asked 6 mins ago
John DoeJohn Doe
1
New contributor
John Doe is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
asked 6 mins ago
John DoeJohn Doe
1
New contributor
John Doe is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
asked 6 mins ago
John DoeJohn Doe
1
New contributor
John Doe is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
asked 6 mins ago
John DoeJohn Doe
1
asked 6 mins ago
John DoeJohn Doe
1
1
New contributor
John Doe is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
New contributor
John Doe is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
John Doe is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
add a comment |
add a comment |
0
active
oldest
votes
Your Answer
StackExchange.ifUsing("editor", function () {
return StackExchange.using("mathjaxEditing", function () {
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
});
});
}, "mathjax-editing");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "557"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: false,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: null,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
John Doe is a new contributor. Be nice, and check out our Code of Conduct.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fdatascience.stackexchange.com%2fquestions%2f46972%2fmask-rcnn-detecting-object-but-mask-is-inaccurate%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
0
active
oldest
votes
0
active
oldest
votes
active
oldest
votes
active
oldest
votes
John Doe is a new contributor. Be nice, and check out our Code of Conduct.
John Doe is a new contributor. Be nice, and check out our Code of Conduct.
John Doe is a new contributor. Be nice, and check out our Code of Conduct.
John Doe is a new contributor. Be nice, and check out our Code of Conduct.
Thanks for contributing an answer to Data Science Stack Exchange!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
Use MathJax to format equations. MathJax reference.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fdatascience.stackexchange.com%2fquestions%2f46972%2fmask-rcnn-detecting-object-but-mask-is-inaccurate%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
