Differences

This shows you the differences between two versions of the page.

--- project:picturethis [2015/02/28 14:24] – [Story] tamberg
+++ project:picturethis [2016/06/21 12:39] – [Software] mariakatosvich
@@ Line 3: / Line 3: @@
 A connected picture frame displaying historic images
-===== Ideas / Iterations =====
+{{https://farm9.staticflickr.com/8681/16672996925_ec53c13c61.jpg}}
-  - Download the pictures to the Raspi and display one of them (warmup)
+===== Story =====
-  - Slideshow and turning the images 90° to adapt to the screensize
-  - Play around with potentiometer and Arduino to bring an analog input onto the Raspi (which only has digital I/O)
-  - Connect everything and adapt the slideshow speed with the potentiometer
-  - Display the name (extracted from the filename) below the picture
-next steps, more ideas:
+  * The Picture This "smart" frame shows police photographs of homeless people by Carl Durheim (1810-1890)
-  - Use the Raspi Cam to detect a visitor in front of the frame and stop the slideshow
+  * By looking at a picture, you trigger a face detection algorithm to analyse both, you and the homeless person
-  - Use the Raspi Cam to take a picture of the face of the visitor
+  * The algorithm detects gender, age and mood of the person on the portrait (not always right)
-  - Detect faces in the images [DONE, manually, using online service]
+  * You, as a spectator, become part of the system / algorithm judging the homeless person
-  - ...merge visitor and picture faces :-)
+  * The person on the picture is at the mercy of the spectator, once again
-===== Story =====
+{{https://farm9.staticflickr.com/8631/16485421338_40e86f1bce.jpg}}
+{{https://farm9.staticflickr.com/8608/16671627211_074399694d.jpg}}
+===== How it works =====
+  * Picture frame has a camera doing face detection for presence detection
+  * Pictures have been pre-processed using a cloud service
+  * Detection is still rather slow (should run faster on the Raspi 2)
+  * Here's a little video https://www.flickr.com/photos/tamberg/16053255113/
+{{https://farm9.staticflickr.com/8608/16672997165_b32d138ee9.jpg}}
+{{https://farm9.staticflickr.com/8577/16671979502_b6a4cc5bd0.jpg}}
+{{https://farm9.staticflickr.com/8611/16673072845_aea4ee6a02_z.jpg}}
+===== Questions (not) answered =====
-  * Picture frame, shows police photographs of the homeless by Carl Durheim (1810-1890)
   * Who were those people? Why were they homeless? What was their crime?
-  * Picture frame has a camera doing face detection, but does not reveal that
-  * Instead, a spectator looking at the frame triggers face detection on the portrait
-  * Face detection algorithm (in the cloud) detects gender, age and mood
-  * Spectator becomes the judge / part of the system / algorithm
-  * Person on the picture is at the mercy of the spectator, once again
   * How would we treat them? How will they / we be treated tomorrow? (by algorithms)
 ===== Data =====
@@ Line 36: / Line 39: @@
   * @tamberg
   * and you
+===== Ideas / Iterations =====
+  - Download the pictures to the Raspi and display one of them (warmup)
+  - Slideshow and turning the images 90° to adapt to the screensize
+  - Play around with potentiometer and Arduino to bring an analog input onto the Raspi (which only has digital I/O)
+  - Connect everything and adapt the slideshow speed with the potentiometer
+  - Display the name (extracted from the filename) below the picture
+next steps, more ideas:
+  - Use the Raspi Cam to detect a visitor in front of the frame and stop the slideshow
+  - Use the Raspi Cam to take a picture of the face of the visitor
+  - Detect faces in the camera picture
+  - Detect faces in the images [DONE, manually, using online service]
+  - ...merge visitor and picture faces :-)
 ===== Material =====
@@ Line 41: / Line 58: @@
   * Raspi https://www.adafruit.com/products/998
   * 7inch TFT https://www.adafruit.com/products/947
+  * Picture frame http://www.thingiverse.com/thing:702589
 ===== Software =====
-  * http://www.raspberrypi.org/downloads/ => Raspbian
+  * http://www.raspberrypi.org/downloads/ - Raspbian
   * http://www.pygame.org - A game framework, allows easy display of images
-  * http://www.raspberrypi.org/facial-recognition-opencv-on-the-camera-board/ - OpenCV on Raspi-Cam
+  * https://bitbucket.org/tamberg/makeopendata/src/tip/2015/PictureThis - Scraper and Display
   * http://skybiometry.com/Demo
+  * http://qwikfix.co.uk/sky-customer-services/
+  * http://simplecv.org
 ===== Links =====
   * https://pinboard.in/search/u:tamberg?query=glamhack
+  * http://www.open-electronics.org/raspberry-pi-and-the-camera-pi-module-face-recognition-tutorial/
+  * http://www.raspberrypi.org/facial-recognition-opencv-on-the-camera-board/ - OpenCV on Raspi-Cam
+  * https://speakerdeck.com/player/083e55006e4a013063711231381528f7 - Slide 106 for face replacement
+  * http://picamera.readthedocs.org/en/release-1.0/
+Not used this time, but might be useful
+  * https://thinkrpi.wordpress.com/2013/05/22/opencv-and-camera-board-csi/ - old Raspian versions did not have the latest OpenCV installed. Nice howto for that
+  * https://realpython.com/blog/python/face-recognition-with-python/ - OpenCV, nice but not used
+  * http://docs.opencv.org - not used directly, only via SimpleCV
+  * http://makezine.com/projects/pi-face-treasure-box/ - maybe a nice weekend project
+  * https://learn.adafruit.com/creepy-face-tracking-portrait
+  * http://www.seeedstudio.com/depot/101LCD-Display-1366x768-HDMIVGANTSCPAL-p-1586.html - 10inch TFT
+  * http://www.aliexpress.com/item/FreeShipping-Banana-Pro-Pi-7-inch-LVDS-LCD-Module-7-Touch-Screen-F-Raspberry-Pi-Car/32246029570.html - 7inch TFT mit LVDS Flex Connector
-{{tag>status:concept needs:dev needs:design needs:data needs:expert glam hardware}}
+{{tag>status:demo glam hardware}}