MVDepthNet
MVDepthNet copied to clipboard
Inaccurate prediction depth
Thank you for your excellent work. Your work is very helpful to me, but I have some problems in depth prediction when using your work. When using example2.py to generate the predicted inverse depth map, although it can effectively distinguish the objects and the background in the scene, there is an estimation error when obtaining the absolute depth of an object. For example, the ground truth is 500mm, but it returns 260mm, while the background is 850mm, but the returned value is 350 mm. Excuse me, what is the problem?
Incidentally, I didn't train my own model, but I used the model you uploaded.
As I am not a major in artificial intelligence, I don't know much about this work. But I used your excellent works in my graduation project, and I am very grateful for your contribution.
As I am not a major in artificial intelligence, I don't know much about this work. But I used your excellent works in my graduation project, and I am very grateful for your contribution.
Hello, can you send me a copy of what you downloaded through the link given by the author? Thank you!And my email address [email protected]
Excuse me, what's the link you're talking about?
------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2022年9月21日(星期三) 下午3:09 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [HKUST-Aerial-Robotics/MVDepthNet] Inaccurate prediction depth (Issue #37)
Hello, can you send me a copy of what you downloaded through the link given by the author? Thank you!
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>
The authors put the two download links above:the model and the sample images. I can't download them.
叫什么名字好呢 @.***
------------------ 原始邮件 ------------------ 发件人: "HKUST-Aerial-Robotics/MVDepthNet" @.>; 发送时间: 2022年9月21日(星期三) 下午3:12 @.>; @.@.>; 主题: Re: [HKUST-Aerial-Robotics/MVDepthNet] Inaccurate prediction depth (Issue #37)
Excuse me, what's the link you're talking about?
------------------ 原始邮件 ------------------
发件人: @.>;
发送时间: 2022年9月21日(星期三) 下午3:09
收件人: @.>;
抄送: @.>; @.>;
主题: Re: [HKUST-Aerial-Robotics/MVDepthNet] Inaccurate prediction depth (Issue #37)
Hello, can you send me a copy of what you downloaded through the link given by the author? Thank you!
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.> — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.>
Could you send me a copy? Thank you very much! I'm having some problems running the code.
叫什么名字好呢 @.***
------------------ 原始邮件 ------------------ 发件人: "HKUST-Aerial-Robotics/MVDepthNet" @.>; 发送时间: 2022年9月21日(星期三) 下午3:12 @.>; @.@.>; 主题: Re: [HKUST-Aerial-Robotics/MVDepthNet] Inaccurate prediction depth (Issue #37)
Excuse me, what's the link you're talking about?
------------------ 原始邮件 ------------------
发件人: @.>;
发送时间: 2022年9月21日(星期三) 下午3:09
收件人: @.>;
抄送: @.>; @.>;
主题: Re: [HKUST-Aerial-Robotics/MVDepthNet] Inaccurate prediction depth (Issue #37)
Hello, can you send me a copy of what you downloaded through the link given by the author? Thank you!
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.> — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.>
You are talking about the best model weight trained by the author, whose name is opensource_model.pth.tar? But I don't know what images you're talking about.
Message ID: @.***>
Yes, I said they were “opensource_model.pth.tar”and“ sample_data.pkl.tar.gz”.The other question is: if I want to run this code, I only need to download these two, right?
叫什么名字好呢 @.***
------------------ 原始邮件 ------------------ 发件人: "HKUST-Aerial-Robotics/MVDepthNet" @.>; 发送时间: 2022年9月21日(星期三) 下午3:29 @.>; @.@.>; 主题: Re: [HKUST-Aerial-Robotics/MVDepthNet] Inaccurate prediction depth (Issue #37)
You are talking about the best model weight trained by the author, whose name is opensource_model.pth.tar? But I don't know what images you're talking about.
Message ID: @.> — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.>
Yes, and the model weights .tar file does not need to be unpacked
I have received it. Thank you very much!
叫什么名字好呢 @.***
------------------ 原始邮件 ------------------ 发件人: "HKUST-Aerial-Robotics/MVDepthNet" @.>; 发送时间: 2022年9月21日(星期三) 下午3:39 @.>; @.@.>; 主题: Re: [HKUST-Aerial-Robotics/MVDepthNet] Inaccurate prediction depth (Issue #37)
Yes, and the model weights .tar file does not need to be unpacked — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>
There is also a question about what the author means by this sentence: “Please download the model via the link and the sample images via the link. Put the model opensource_model.pth.tar and extract the sample_data.pkl.tar.gz under the project folder.”
叫什么名字好呢 @.***
------------------ 原始邮件 ------------------ 发件人: "HKUST-Aerial-Robotics/MVDepthNet" @.>; 发送时间: 2022年9月21日(星期三) 下午3:39 @.>; @.@.>; 主题: Re: [HKUST-Aerial-Robotics/MVDepthNet] Inaccurate prediction depth (Issue #37)
Yes, and the model weights .tar file does not need to be unpacked — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>
This means you need to put those two downloaded files in the same folder as example.py and example2.py,then run example.py
Message ID: @.***>
Ok, thank you very much!
叫什么名字好呢 @.***
------------------ 原始邮件 ------------------ 发件人: "HKUST-Aerial-Robotics/MVDepthNet" @.>; 发送时间: 2022年9月21日(星期三) 下午4:00 @.>; @.@.>; 主题: Re: [HKUST-Aerial-Robotics/MVDepthNet] Inaccurate prediction depth (Issue #37)
This means you need to put those two downloaded files in the same folder as example.py and example2.py,then run example.py
Message ID: @.> — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.>
Thank you. Can you send me a copy?
叫什么名字好呢 @.***
------------------ 原始邮件 ------------------ 发件人: "HKUST-Aerial-Robotics/MVDepthNet" @.>; 发送时间: 2022年9月21日(星期三) 下午3:39 @.>; @.@.>; 主题: Re: [HKUST-Aerial-Robotics/MVDepthNet] Inaccurate prediction depth (Issue #37)
Yes, and the model weights .tar file does not need to be unpacked — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>
I sent it to you by e-mail before. Didn't you get it?
Message ID: @.***>
I have received,Thank you very much
叫什么名字好呢 @.***
------------------ 原始邮件 ------------------ 发件人: "HKUST-Aerial-Robotics/MVDepthNet" @.>; 发送时间: 2022年10月11日(星期二) 下午3:05 @.>; @.@.>; 主题: Re: [HKUST-Aerial-Robotics/MVDepthNet] Inaccurate prediction depth (Issue #37)
I sent it to you by e-mail before. Didn't you get it?
Message ID: @.> — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.>
You are talking about the best model weight trained by the author, whose name is opensource_model.pth.tar? But I don't know what kind of picture you're talking about.
Message ID: @.***>
I'm sorry. I don't know why I sent you the same message as last time half an hour ago. It's my fault. Thank you for your reply.
叫什么名字好呢 @.***
------------------ 原始邮件 ------------------ 发件人: "HKUST-Aerial-Robotics/MVDepthNet" @.>; 发送时间: 2022年10月11日(星期二) 下午3:29 @.>; @.@.>; 主题: Re: [HKUST-Aerial-Robotics/MVDepthNet] Inaccurate prediction depth (Issue #37)
You are talking about the best model weight trained by the author, whose name is opensource_model.pth.tar? But I don't know what kind of picture you're talking about.
Message ID: @.> — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.>
@979362835 Hi! Can you send me the model and images too? I can't download them ( [email protected]
Thank you!
What you asked for has been sent to your mailbox.
@979362835 Got it. Thank you!
@979362835 Hi, I would like to get model files also. Thank you.
@979362835 Hi, I would like to get model files also. Thank you.
You can leave your email address and I will give you the model the next day.
@979362835 Hi, I would like to get model files and samples, since I can't download them through baidu. My email is [email protected].
@979362835 Hi, I would like to get model files also. Thank you.
You can leave your email address and I will give you the model the next day.
My email is: [email protected]
@979362835 Hi, I would like to get model files and samples, since I can't download them through baidu. My email is [email protected].
I have already sent it.
Thank you sir!
@979362835 Hi, I would like to get model files also. Thank you.
You can leave your email address and I will give you the model the next day.
My email is: [email protected]
I have already sent it.
Thanks alot