Differences

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

Link to this comparison view

Both sides previous revision Previous revision
Next revision		 Previous revision
lirec:faceident_dev_log [2009-05-27 11:27] – davegriffithslirec:faceident_dev_log [2009-05-27 11:53] (current) – davegriffiths
Line 40: Line 40:
 Added a simple message to tell other programs which user is in view, with a confidence metric. Added a simple message to tell other programs which user is in view, with a confidence metric.
  
-^ /faceident user-id confidence-value ^+| /faceident user-id confidence-value |
  
 The program still needs to be driven by user input, calibrating it by recording people at startup. The program still needs to be driven by user input, calibrating it by recording people at startup.
 +
 +====Scene state [32rd March 2009]====
 +
 +Rather than simply outputting the user id's every frame, it's more useful if the program tracks the state of the scene internally, and outputs changes. I've added the following yarp messages:
 +
 +^ Output port  ^   Bottle contents ^                Meaning ^
 +| /faceident |    "user appeared" ID confidence |  A user has entered the view of the camera |
 +| /faceident |    "user disappeared" ID          | The user has left the camera view |
  
 ====More YARP, and persistence for INESC-ID [15 April 2009]==== ====More YARP, and persistence for INESC-ID [15 April 2009]====
Line 51: Line 59:
 | /faceident-ctrl | "train" ID    |      Train for this user | | /faceident-ctrl | "train" ID    |      Train for this user |
 | /faceident-ctrl | "detect"       |     Switch out of training mode | | /faceident-ctrl | "detect"       |     Switch out of training mode |
-| /faceident-ctrl | "save" sessionname |  Save the detected faces  +| /faceident-ctrl | "save" sessionname |  Save the detected faces | 
-| /faceident-ctrl | "load" sessionname | Load previously detected faces +| /faceident-ctrl | "load" sessionname | Load previously detected faces |
 | /faceident-ctrl | "clear"          |   Clears all faces | | /faceident-ctrl | "clear"          |   Clears all faces |
  
Line 60: Line 68:
 would be to lower the error threshold and sneakily recalibrate for the most likely user at the start of a new session. would be to lower the error threshold and sneakily recalibrate for the most likely user at the start of a new session.
  
-====+====Idle mode [15 April 2009]==== 
 + 
 +Due to a request from INESC-ID, which helps thinking about this more in terms of the wider architecture, it's important that competencies can be tured on or off easily, particularly if they use up cpu, as computer vision competencies will. To this end I've added an idle command: 
 + 
 +^ Input port    ^  Bottle contents  ^   Meaning ^ 
 +| /faceident-ctrl | "idle" | Switch to idle mode, mostly frees up cpu | 
 + 
 +====XML and experimental multi image mode [23 April 2009]==== 
 + 
 +Using [[http://www.grinninglizard.com/tinyxml/|TinyXML]] as a source library for making the persistence much more scalable and user editable.  
 + 
 +I've also added another yarp message: 
 +^ Input port    ^  Bottle contents  ^   Meaning ^ 
 +| /faceident-ctrl | "multiimages" 1/0   |   Turn on or off multiple images per face, experimental | 
 + 
 +This is the first step to dealing with lighting changes, just allowing images of people in multiple lighting situations to be used at once. The system looks through each image for each person in order to match the closest. 
 + 
 +In calibration mode, the system will add new images to the user if the detected ones are above the error threshold. This is slightly dangerous, as you would be able to get images for different people in the same id... Not sure if there is a way to cope with that automatically. 
  
  
  • lirec/faceident_dev_log.1243423633.txt.gz
  • Last modified: 2009-05-27 11:27
  • by davegriffiths