skip to navigation
skip to content

Not Logged In

collective.takeaportrait 0.2.0

Plone site members can change personal portrait using a webcam

This product use super-modern browser feature for accessing the user's webcam (if any) to directly save a new portrait photo inside Plone personal preferences.

How it works

Just add the product to your buildout and activate the collective.takeaportrait add-on.

After that you will probably see a new button inside your "Personal Information" labeled "Take a photo". If you don't see any new button your browser will probably not support Media Capture API.

alternate text alternate text alternate text

When pressing the button the browser will ask you the permission to access the Webcam and you must accept to continue. The security request format is vendor specific.

When confirmed you will see the new portrait overlay, where the webcam is activated and the output stream is directly on your browser.

alternate text

You have two possible actions: close the window or take a photo. In the latter case a delay counter will be displayed on the left (you can raise or lower the delay using the range control, from 0 to 10 seconds).

alternate text

Note: only the image part inside the highlight yellow section (the viewfinder) will be used as a new portrait. You can move the viewfinder by drag and drop with the mouse (not working on mobile right now).

After every shot taken, you can look at a preview of the image, and repeat the operation until you get a nice portrait.

alternate text

Finally, you can save the new portrait in the user's preferences. The image in the form will be immediately updated from the server.

alternate text

Face detection

The product contains a face detection support, that can automatically center the viewfinder on any recognized face.

Default implementation rely onto the OpenCV, Computer Vision Library (that must be installed on your Python env) and must be explicitly activated:

[instance]
...
zcml=
    ...
    collective.takeaportrait.opencv

Note

This feature in deeply experimental, a proof-of-concept not ready from production environment.

See Plone, HTML 5 Canvas and Face Detection with Webcam article for more.

If you manually move the viewfinder, the auto-center feature based on face detection will be stopped.

Plone integration

As customizing the user preference form is something I don't like very much, this product is adding new features only using pure JavaScript.

Tested on Plone 4.3.

Browsers support

The user's browser must support HTML Media Capture API so only recent and cool browser can be used:

  • Firefox: tested with Firefox 25
  • Chrome: tested with Chrome 30 (a little slow, but works)
  • Opera: tested with Opera Next, version 18
  • Safari: no support (c'mon Apple...)
  • Internet Explorer: support is expected for Internet Explorer 31 [1]

To know if your browser is supported, see the Can I Use Web site.

Mobile

  • Safari on iOS: no support
  • Android Browser (Android 4.2.2): no support
  • Chrome on Android: tested with Chrome 32
  • Firefox: on version 27 camera seems to work but something goes wrong
[1]Microsoft said that probably the Media Capture support will be delayed to Internet Explorer 36 in case Patent War VI against Kilrathi race on Tau Ceti will not over on time.

Changelog

0.2.0 (2014-02-20)

  • Viewfinder can now be dragged for optimal face position [keul]
  • Face detection support (super-experimental) [keul]

0.1.0 (2013-11-10)

  • Initial release
 
File Type Py Version Uploaded on Size
collective.takeaportrait-0.2.0.zip (md5) Source 2014-02-20 140KB
  • Downloads (All Versions):
  • 34 downloads in the last day
  • 250 downloads in the last week
  • 597 downloads in the last month