» DetectObject (FreeFrame DShow9)
This site relies heavily on Javascript. You should enable it if you want the full experience. Learn more.

DetectObject (FreeFrame DShow9)

Help
Author
Info This node comes with the addon pack. Get it here.

Here you are supposed to see an image (rendered with SVG) describing the pins of this node. However, this does not work with Internet Explorer. You may want to consider using another browser, like FireFox, Chrome, Safari or Opera that do support SVG.

Credits
Warnings
Bugs

related discussions

start a forum thread about this node

about DetectObject (FreeFrame DShow9)

open source

usage

DetectObject is simply a wrapper around the openCV function cvHaarDetectObjects.

DetectObject finds rectangular regions in the given image that are likely to contain objects the cascade has been trained for and returns those regions as a sequence of rectangles. DetectObject scans the image several times at different scales (-> Scale Cascade). Each time it considers overlapping regions in the image and applies the classifiers to the regions. It may also apply some heuristics to reduce number of analyzed regions, such as Canny pruning. After it has proceeded and collected the candidate rectangles (regions that passed the classifier cascade), it groups them and returns a sequence of average rectangles for each large enough group. The default parameters:

  • Scale Cascade = 0.1
  • Min Neighbours = 3
  • Canny Pruning = 0

are tuned for accurate yet slow face detection. For faster face detection on real video images the better settings are:

  • Scale Cascade = 0.2
  • Min Neighbours = 2
  • Canny Pruning = 1

text adopted from:
Emgu CV Library Documentation

Training Files

training files and an application to create those files are shipped with openCVhttp://sourceforge.net/projects/opencvlibrary

there is a wiki with some haarcascade training files for different purposes.

Links

OpenCVwiki

release info

  • 17 06 07
    • added Unique IDs and Lifetime pins (development sponsored by dotdotdot)

anonymous user login

Shoutbox

~14h ago

antokhio: dxt slow as hell, but I like that quality alot :;]

~18h ago

vux: @antokhio there's some benefits from hap over dxt1/5 image sequence (extra compression+data locality)

~19h ago

ggml: not sure what you mean.. new thread ?

~20h ago

graphicuserinterface: @antokhio, so we had years ago a similar tech, only not applied to video?

~20h ago

defetto: hehehe are you questioning the benefits of visual programming over text? ;)

~20h ago

microdee: @defetto: learning HLSL can save you 90 bucks ;)

~22h ago

circuitb: @anto Hap Alpha should be DXT5?

~22h ago

antokhio: hap = dxt1 dds seq

~1d ago

ggml: @antokhio, hap: how does this relate to the question about a possible hap release ?