Giter Club home page Giter Club logo

opencv-rgbdodometry's Introduction

Hi there ๐Ÿ‘‹

I am TzuTa. You can call me Darren.

TzuTa's github stats Top Langs

  • ๐Ÿ”ญ Iโ€™m currently working for Amazon to build big data solution, recommender algorithm, Tier-1 services, web application and mobile app. After I joined Amazon, I didn't contribute to open source anymore. But if you have any question or would like to know more about me. You can check out https://tzutalin.github.io to reach out to me

opencv-rgbdodometry's People

Contributors

tzutalin avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

opencv-rgbdodometry's Issues

opencv 2.4.10 misses utility.hpp

Hello,

I am trying to build it with a VS2010 on Windows 7, with which the OpenCV installed is only 2.4.10.
The compiler gives an error that core/utility.hpp cannot found. I downloaded one utility.hpp from OpenCV's repository but doesn't work.
How it is possible to make this program working with OpenCV 2.4.10. Thanks.

Getting zero correspondences with ICP based odometry

I am trying with other type of odometry, Rgbd-ICP and ICP. Its not working, unable to find correspondences, I changed main code accordingly as shown below:

odom = std::make_shared<rgbd::RgbdICPOdometry>(
          cameraMatrix, MIN_DEPTH, MAX_DEPTH, MAX_DEPTH_DIFF, MAX_POINTS_PART, iterCounts,
          minGradMagnitudes, rgbd::Odometry::RIGID_BODY_MOTION);
      /* 
      odom = std::make_shared<rgbd::ICPOdometry>(
          cameraMatrix, MIN_DEPTH, MAX_DEPTH, MAX_DEPTH_DIFF, MAX_POINTS_PART, iterCounts,
          rgbd::Odometry::RIGID_BODY_MOTION);
      */

Run with Gazebo Simulation.

Hi, this code works perfect with the TUM dataset but I want to run it with images obtained from a Gazebo Simulation.
To get the rgb images I use the topic /camera/rgb/image_color .
And for the depth: /camera/depth/image. This topic gets depth maps in meters directly, so I think that don't need to make any conversion.

The Gazebo kinect model has the same calibration parameters that in the code. Also I think that I have created a gazebo world with enough texture.

I thought that it may work well but no. I don't know where is the problem. When I run it appears rgb and depth images well but, the trajectory is not correct.

Do you know if I have to make any image conversion or something?

Thank you in advance.

RgbdOdometry with Xtion

Hi, i'm trying to integrate the library with OpenNI and Xtion sensor, but compute always fail without any message, please advice the most critical places in lib code to debug.

	// in what units should be focus and CX/CY?
	float vals[] = {FOCUS_LENGTH, 0., CX, 0., FOCUS_LENGTH, CY, 0., 0., 1.};

	new rgbd::RgbdOdometry() // like in example
	
	// loop..
	
	auto depth = oni_sensor_->get_depth_fd();
	auto color = oni_sensor_->get_color_fd();
	
	cv::Mat depth_16u(depth.height, depth.width, CV_16U, (char*)depth.data);
	cv::Mat depth_image;
	depth_16u.convertTo(depth_image, CV_32FC1, 0.001f); // mm -> m

	cv::Mat color_image(color.height, color.width, CV_8UC3, (char*)color.data);
	cv::Mat gray_image;
	cvtColor(color_image, gray_image, cv::COLOR_RGB2GRAY);

	cv::Mat normalizeDepth;
	depth_image.convertTo(normalizeDepth, CV_8UC1, 255.0f / 4.0f);
	cv::imshow("x0", normalizeDepth);
	cv::imshow("x1", gray_image);
	
	bool res = odom_->compute(image0_, depth0_, cv::Mat(), gray_image, depth_image, cv::Mat(), Rt);

	// next time use current frame as source	
	depth_image.copyTo(depth0_.get);
	gray_image.copyTo(image0_.get);

test_frame2

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.