jkff00 / PointCloudColorizeDemo

This is a project demo that creating a virtual scene by the Dual-LIDAR&&multi-fisheyes camera-rig
2 stars 0 forks source link

I want to Know your colorize funciton? #1

Open Why384040146 opened 11 months ago

Why384040146 commented 11 months ago

Hi, I found your color pointcloud so beautiful, can you suggest your colorized idea ? tks

jkff00 commented 11 months ago

In short, the color difference of the image needs to be solved, followed by high-precision camera pose estimation, and finally evaluating the quality of each pixel (through the Angle between the pixel ray and the optical axis and the depth) to determine which pixels are fine enough, and that's about all we did.

---Original--- From: @.> Date: Sun, Oct 8, 2023 20:40 PM To: @.>; Cc: @.***>; Subject: [jkff00/PointCloudColorizeDemo] I want to Know your colorizefunciton? (Issue #1)

Hi, I found your color pointcloud so beautiful, can you suggest your colorized idea ? tks

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: @.***>

Yellowshuohahaha commented 8 months ago

@jkff00 Hi, the project is an open source project?

jkff00 commented 8 months ago

Thanks for your question.The project is still ongoing and may not be open-sourced in the short term. The following graphic describes how the color quality metric is calculated, with w being a custom weight, and the angle and depth of a single 3D point being observed by multiple views normalized to calculate the quality score of that point in each image. Obviously, the smaller the score the higher the quality. The coloring blends the RGB information of some of the views according to personal needs is sufficient.

WINNER 东区 @.***

 

------------------ 原始邮件 ------------------ 发件人: "jkff00/PointCloudColorizeDemo" @.>; 发送时间: 2024年1月2日(星期二) 中午11:35 @.>; 抄送: "WINNER @.**@.>; 主题: Re: [jkff00/PointCloudColorizeDemo] I want to Know your colorize funciton? (Issue #1)

@jkff00 Hi, the project is an open source project?

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: @.***>

Why384040146 commented 5 months ago

Thanks for your question.The project is still ongoing and may not be open-sourced in the short term. The following graphic describes how the color quality metric is calculated, with w being a custom weight, and the angle and depth of a single 3D point being observed by multiple views normalized to calculate the quality score of that point in each image. Obviously, the smaller the score the higher the quality. The coloring blends the RGB information of some of the views according to personal needs is sufficient. WINNER 东区 @.   ------------------ 原始邮件 ------------------ 发件人: "jkff00/PointCloudColorizeDemo" @.>; 发送时间: 2024年1月2日(星期二) 中午11:35 @.>; 抄送: "WINNER @*.**@*.>; 主题: Re: [jkff00/PointCloudColorizeDemo] I want to Know your colorize funciton? (Issue #1) @jkff00 Hi, the project is an open source project? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: @.>

excellent work, expect your open source code...

Why384040146 commented 2 months ago

Thanks for your question.The project is still ongoing and may not be open-sourced in the short term. The following graphic describes how the color quality metric is calculated, with w being a custom weight, and the angle and depth of a single 3D point being observed by multiple views normalized to calculate the quality score of that point in each image. Obviously, the smaller the score the higher the quality. The coloring blends the RGB information of some of the views according to personal needs is sufficient. WINNER 东区 @.   ------------------ 原始邮件 ------------------ 发件人: "jkff00/PointCloudColorizeDemo" @.>; 发送时间: 2024年1月2日(星期二) 中午11:35 @.>; 抄送: "WINNER @*.**@*.>; 主题: Re: [jkff00/PointCloudColorizeDemo] I want to Know your colorize funciton? (Issue #1) @jkff00 Hi, the project is an open source project? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: @.>

Thanks for your question.The project is still ongoing and may not be open-sourced in the short term. The following graphic describes how the color quality metric is calculated, with w being a custom weight, and the angle and depth of a single 3D point being observed by multiple views normalized to calculate the quality score of that point in each image. Obviously, the smaller the score the higher the quality. The coloring blends the RGB information of some of the views according to personal needs is sufficient. WINNER 东区 @.   ------------------ 原始邮件 ------------------ 发件人: "jkff00/PointCloudColorizeDemo" @.>; 发送时间: 2024年1月2日(星期二) 中午11:35 @.>; 抄送: "WINNER @*.**@*.>; 主题: Re: [jkff00/PointCloudColorizeDemo] I want to Know your colorize funciton? (Issue #1) @jkff00 Hi, the project is an open source project? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: @.>

I canot see you graphic descirbes this weight para, can you post the graph ?