DELPHI OPENCV : SURF detection against a database - database

I am an OpenCv newbie,and i just want to built something like this example :
OPENCV: SURF detection against a database of 20000 images
I just want to know what are the steps to follow to achieve this, i managed to integrate OpenCv 2.4.13 with Delphi Berlin 10.1.
It seems to be working i even tried some examples that i found here : https://github.com/Laex/Delphi-OpenCV
Basically what i wanna do is realtime detection against a database, using a webcam as shown in the previous video.

Related

Implementing "Night Mode" in GIS with WMS

I am fairly new to the world of developing and the world of mapping. For my studies I want to implement a WMS map into QGIS/ArcGIS but have it in a night mode/dark mode style for better visual representation. Now as I looked through the web, I did not find a service or provider who has a WMS server with a night mode OSM (Open street map). Does somebody of you have an idea on how to solve this problem? Maybe even with a python code?
Thanks in advance!!!
If you are using leaflet in JS or QGIS for example you can use xyz tiles. Here you can find some examples.
For QGIS, get the link that looks like 'https://example.com/{z}/{x}/{y}.png' in the plane JavaScript code snippet in the website and add a new xyz connection.

How to link a 3ds max scene with my directx 9 game?

I have purchased the book programming a multiplayer fps book.
The problem is that I am not able to get the 3d's max scene to load. Is there a step to load a library of 3d's max to load the scene? What is the missing piece in loading the scene. I am not able to load the scene.
In order to load and render 3D mesh data, you need some kind of exporter to convert the 3D Studio MAX authoring data into a binary form suitable for rendering, and some sort of rendering engine capable of drawing that binary form.
For exporting the data, you can use an existing tool such as Open-Asset-Importer-Lib or write your own using something like Autodesk FBX.
There are of course numerous choices for rendering engines, or you can write your own from scratch.
Note that Direct3D 9 is extremely old at this point. DirectX 11 is the modern sweet spot for starting out programming Direct3D. See DirectX Tool Kit and DirectX SDK Samples Content Exporter.
The 3dsmax file format is propriatary. You just don't load those *.max files into anything but 3ds max. Like Chuck said above, you have to export your max file from 3ds max into a common, 3rd party file format like OBJ, FBX, etc...
There is probably a hundred 3D third-party file formats out there. We have no idea which 'game' you are trying to code up, and you have not provided any information, except asked all the wrong questions.
DirectX 11 is a graphics platform provided by microsoft. it is file format agnostic. It is up to you to learn it.

choosing 3D library / platform for mobile and Desktop application

I want to build an application for mobile (ios,android) and Desktop (Windows) or Web.
The application will look like this: a 3D object which the user can play with the camera perspective around it and some menus.
What I need is to manipulate a 3D Object like torus or tube. by manipulate I mean: change materials and edit the object like a polygon. of course I also need menus and communication to a server.
*Optional: I am not sure if I need to load 3D model from a file.
What I don't understand is should I look for all in one solution or combine a cross platform framework with other libraries? Are Game engine suitable for this task?
My options so far:
Use Three.js with PhoneGap and write in java script
Use OGRE and write in C++ with some cross-platform framework that allow me to write in C++.
I never developed for mobile and I wonder about those cross-platform frameworks: does the application's size is big? does the application runs well? which obstacles should i have comparing writing separate applications for each platform?
Thank you for your help
Might want to look at libgdx .

How to get a picture from a kinect with labview

My class and I are trying to make a robot, we need to be able to take a picture using a kinect in Labview. We know how to get a skeleton from the kinect, but we can't figure out how to take a picture/video. Is there a DLL we need to download? We can't find anything on the internet but we know it's possible to do in labview. So do we need to write any code in C? Im pretty good with C so don't restrict your answer to just labview. All relevant answers are welcome.
We can't find anything on the internet but we know it's possible to do
in labview.
Personally, I have no experience with this, but a quick "LabVIEW Kinect" search on Google turns up any number of results, including some which show video (example). Based on the videos and images I see there, it looks like they're using the IMAQ control to show the video, so I'm assuming you're going to need the vision toolkit for that. If you have an academic license, you might already have that, but even if not, you could try asking your local NI office for one, as they generally tend to support schools.
I would go on ni.com and search for LabVIEW Kinect. There are links over there like this:
Kinect LabVIEW Interface Using Microsoft Kinect API

How to rotate image in android using android ndk

i want to rotate bitmap in android . i don't want to use android api but want to use c code for image processing and fast response. can u tell me how to achieve this in c??
Thank you in advance
You could use OpenCV for Android by processing a rotation matrix in native code. But if you really just want to rotate an image, using native code for this would be over the top, unless you plan adding more complex transformations. A single transformation supposedly won't be processed faster in comparison to just rotating it in Java.
There is a nice solution for your needs: ImageMagick. You can compile C version of ImageMagick with NDK. There is also ImageMagick-Android project on github, but it doesn't look like a ready-to-use solution.

Resources