Is it possible to view what Q.bo sees with the camera's? I tried some webcam software on the Raspberry but they cannot access the camera's.
Hi Peter,
Definitely. You can use "fswebcam".
First, install the fswebcam package:
sudo apt-get install fswebcamEnter the command fswebcam followed by a filename and a picture will be taken using the webcam, and saved to the filename specified:
fswebcam image.jpgThis command will show the following information:
--- Opening /dev/video0...
Trying source module v4l2...
/dev/video0 opened.
No input was specified, using the first.
Adjusting resolution from 384x288 to 352x288.
--- Capturing frame...
Corrupt JPEG data: 2 extraneous bytes before marker 0xd4
Captured frame in 0.00 seconds.
--- Processing captured image...
Writing JPEG image to 'image.jpg'.
That worked!
It's planned any tool to view in real time what's qbo viewing?
Besides the usefulness of the tool itself (having a look upon you children, control what's happening at your home when you're out, etc) it will help to see if illumination is good enough for qbo to recognice you, etc.
That worked!
It's planned any tool to view in real time what's qbo viewing?
Besides the usefulness of the tool itself (having a look upon you children, control what's happening at your home when you're out, etc) it will help to see if illumination is good enough for qbo to recognice you, etc.
Hi Malevolent,
yes, we are working on it. We will release this new feature in the 3rd Version of OpenQbo.
When I tried this:
fswebcam image1.jpg
--- Opening /dev/video0...
Trying source module v4l2...
/dev/video0 opened.
No input was specified, using the first.
Error selecting input 0
VIDIOC_S_INPUT: Device or resource busy
what is happening?
When I tried this:
fswebcam image1.jpg
--- Opening /dev/video0...
Trying source module v4l2...
/dev/video0 opened.
No input was specified, using the first.
Error selecting input 0
VIDIOC_S_INPUT: Device or resource busywhat is happening?
you need to "kill" all processes running in real time ( tracking, recognition, etc )
@greybeard81: did above answer help you to resolve your problem?
Thanks for posting how to capture an image of what Q.bo is looking at. I noticed that when put on a table it looked at the ceiling instead of the head of a person of normal height. I thought it didn't work initially. Great tool!
--
Christian Buysschaert
Leuven - Belgium
Email christian.buysschaert@gmail.com
Not really. Although my camera must have been working while setting up my wireless network reading the code symbol, it has never tracked anything. I changed the token to mine for interactive and only the ears respond. No tracking. I went to scratch and loaded the demo program and still no tracking. All the other features work, up, down, left, right, and hitting 'a ' made it go through all the speech patterns. I have been trying a variety of dialogflow agents to no avail including Smalltalk. I have never seen a green or blue led nose indicating that it is listening. Help!
Not really. Although my camera must have been working while setting up my wireless network reading the code symbol, it has never tracked anything. I changed the token to mine for interactive and only the ears respond. No tracking. I went to scratch and loaded the demo program and still no tracking. All the other features work, up, down, left, right, and hitting 'a ' made it go through all the speech patterns. I have been trying a variety of dialogflow agents to no avail including Smalltalk. I have never seen a green or blue led nose indicating that it is listening. Help!
Maybe this can help you:
Open the base robot ( remove the two white plastic screws ). Unplug the usb camera and connect the other free usb instead.
Apparently, it worked. I expected qbo to look for a face but I finally got a response for a face by rotating the head back and forth and up and down to finally locate a face. I am disappointed by the long latency for a verbal response but it finally worked. The only failure was a response to 'bye bye'. The right ear responded to my change in the dialog. I plan to keep working on how to program qbo. It would be nice if there were instructions on how to program the various motions and responses, mouth, nose etc. I have had no luck with dialogflow so far. Any help would be appreciated. How about a qbo program listed on dialogflow's home page among the other programs listed. I am going the long haul to tame this beast!
Apparently, it worked. I expected qbo to look for a face but I finally got a response for a face by rotating the head back and forth and up and down to finally locate a face. I am disappointed by the long latency for a verbal response but it finally worked. The only failure was a response to 'bye bye'. The right ear responded to my change in the dialog. I plan to keep working on how to program qbo. It would be nice if there were instructions on how to program the various motions and responses, mouth, nose etc. I have had no luck with dialogflow so far. Any help would be appreciated. How about a qbo program listed on dialogflow's home page among the other programs listed. I am going the long haul to tame this beast!
Hahaha. I like "I am going the long hau to tame this beast!" ;)
The latency is due to the location of the servers. You can try to install Google Assistance into Q.bo and see how Q.bo respond in "real time".
I've written a small Python script that will show you a realtime video what Q.bo sees and when/which faces he detects by drawing colored rectangles. Green rectangles are detected frontal faces, blue rectangles are detected profile faces. Have fun!
--
Christian Buysschaert
Leuven - Belgium
Email christian.buysschaert@gmail.com
When I tried this:
fswebcam image1.jpg
--- Opening /dev/video0...
Trying source module v4l2...
/dev/video0 opened.
No input was specified, using the first.
Error selecting input 0
VIDIOC_S_INPUT: Device or resource busywhat is happening?
you need to "kill" all processes running in real time ( tracking, recognition, etc )
How does one "kill" all processes? I am having the same error.
Apparently, it worked. I expected qbo to look for a face but I finally got a response for a face by rotating the head back and forth and up and down to finally locate a face. I am disappointed by the long latency for a verbal response but it finally worked. The only failure was a response to 'bye bye'. The right ear responded to my change in the dialog. I plan to keep working on how to program qbo. It would be nice if there were instructions on how to program the various motions and responses, mouth, nose etc. I have had no luck with dialogflow so far. Any help would be appreciated. How about a qbo program listed on dialogflow's home page among the other programs listed. I am going the long haul to tame this beast!
Hahaha. I like "I am going the long hau to tame this beast!" ;)
The latency is due to the location of the servers. You can try to install Google Assistance into Q.bo and see how Q.bo respond in "real time".
So I got this error "File "Qbo_webcam.py", line 40, in <module>
webcam = cv2.VideoCapture(int(config['camera'])) # Get ready to start getting images from the webcam
TypeError: 'NoneType' object has no attribute '__getitem__' "
