I want use new videomapping

Question

I'd like to use SaveVideo module in a different size.

so, I wrote NONE 0 0 0 YUYV 640 480 30.0 JeVois SaveVideo in jevois/config/videomappings.cfg

then, I excuted sudo jevois-add-videomapping NONE 0 0 0 YUYV 640 480 30.0 JeVois SaveVideo

and checked listmappings

AVAILABLE VIDEO MAPPINGS:

    0 - OUT: NONE 0x0 @ 0fps CAM: YUYV 1280x1024 @ 15fps MOD: JeVois:DemoARtoolkit C++
    1 - OUT: NONE 0x0 @ 0fps CAM: YUYV 640x480 @ 30fps MOD: JeVois:SaveVideo C++
    2 - OUT: NONE 0x0 @ 0fps CAM: YUYV 640x480 @ 30fps MOD: JeVois:DemoARtoolkit C++
    3 - OUT: NONE 0x0 @ 0fps CAM: YUYV 320x240 @ 60fps MOD: JeVois:DemoARtoolkit C++
    4 - OUT: NONE 0x0 @ 0fps CAM: YUYV 320x240 @ 60fps MOD: JeVois:ObjectTracker C++
    5 - OUT: NONE 0x0 @ 0fps CAM: YUYV 320x240 @ 60fps MOD: JeVois:FirstVision C++
    6 - OUT: NONE 0x0 @ 0fps CAM: YUYV 320x240 @ 60fps MOD: JeVois:FirstPython Python
    7 - OUT: NONE 0x0 @ 0fps CAM: YUYV 320x240 @ 60fps MOD: JeVois:SaveVideo C++
    8 - OUT: NONE 0x0 @ 0fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:SaveVideo C++
    9 - OUT: NONE 0x0 @ 0fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:RoadNavigation C++
   10 - OUT: NONE 0x0 @ 0fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:DemoArUco C++
   11 - OUT: NONE 0x0 @ 0fps CAM: YUYV 320x240 @ 15fps MOD: JeVois:PythonObject6D Python
   12 - OUT: NONE 0x0 @ 0fps CAM: YUYV 176x144 @ 120fps MOD: JeVois:FirstVision C++
   13 - OUT: NONE 0x0 @ 0fps CAM: YUYV 176x144 @ 120fps MOD: JeVois:SaveVideo C++
   14 - OUT: NONE 0x0 @ 0fps CAM: YUYV 176x144 @ 120fps MOD: JeVois:FirstPython Python
   15 - OUT: 0�� 32767x3 @ 0fps CAM: \s+ 640x480 @ 60fps MOD: JeVois:SaveVideo C++
   16 - OUT: YUYV 1280x1092 @ 7fps CAM: YUYV 1280x1024 @ 7fps MOD: JeVois:TensorFlowEasy C++
   17 - OUT: YUYV 1280x480 @ 15fps CAM: YUYV 640x480 @ 15fps MOD: JeVois:DarknetYOLO C++
   18 - OUT: YUYV 960x240 @ 30fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:DemoNeon C++
   19 - OUT: YUYV 880x480 @ 10fps CAM: YUYV 640x480 @ 10fps MOD: JeVois:DarknetSaliency C++
   20 - OUT: YUYV 640x546 @ 20fps CAM: YUYV 640x480 @ 20fps MOD: JeVois:MarkersCombo C++
   21 - OUT: YUYV 640x500 @ 20fps CAM: YUYV 640x480 @ 20fps MOD: JeVois:DemoArUco C++
   22 - OUT: YUYV 640x480 @ 60fps CAM: YUYV 640x480 @ 60fps MOD: JeVois:SaveVideo C++
   23 - OUT: YUYV 640x480 @ 30fps CAM: YUYV 640x480 @ 30fps MOD: JeVois:SaveVideo C++
   24 - OUT: YUYV 640x480 @ 29fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:JeVoisIntro C++
   25 - OUT: YUYV 640x480 @ 15fps CAM: YUYV 640x480 @ 15fps MOD: JeVois:PythonTest Python
   26 - OUT: YUYV 640x480 @ 7.5fps CAM: YUYV 640x480 @ 7.5fps MOD: JeVois:DiceCounter C++
   27 - OUT: YUYV 640x360 @ 30fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:JeVoisIntro C++
   28 - OUT: YUYV 640x312 @ 50fps CAM: YUYV 320x240 @ 50fps MOD: JeVois:DemoSalGistFaceObj C++
   29 - OUT: YUYV 640x300 @ 60fps CAM: YUYV 320x240 @ 60fps MOD: JeVois:DemoSaliency C++
   30 - OUT: YUYV 640x290 @ 60fps CAM: YUYV 320x240 @ 60fps MOD: JeVois:FirstVision C++
   31 - OUT: YUYV 640x252 @ 60fps CAM: YUYV 320x240 @ 60fps MOD: JeVois:FirstPython Python
   32 - OUT: YUYV 640x240 @ 30fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:ColorFiltering C++
   33 - OUT: YUYV 544x240 @ 15fps CAM: YUYV 320x240 @ 15fps MOD: JeVois:DarknetSingle C++
   34 - OUT: YUYV 512x240 @ 30fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:TensorFlowSaliency C++
   35 - OUT: YUYV 464x240 @ 15fps CAM: YUYV 320x240 @ 15fps MOD: JeVois:TensorFlowSingle C++
   36 - OUT: YUYV 352x288 @ 30fps CAM: YUYV 352x288 @ 30fps MOD: JeVois:PythonSandbox Python
   37 - OUT: YUYV 320x308 @ 30fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:TensorFlowEasy C++
   38 - OUT: YUYV 320x290 @ 60fps CAM: YUYV 320x240 @ 60fps MOD: JeVois:FirstVision C++
   39 - OUT: YUYV 320x288 @ 30fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:SaliencySURF C++
   40 - OUT: YUYV 320x286 @ 30fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:DemoQRcode C++
   41 - OUT: YUYV 320x266 @ 30fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:ArUcoBlob C++
   42 - OUT: YUYV 320x262 @ 15fps CAM: YUYV 320x240 @ 15fps MOD: JeVois:PythonObject6D Python
   43 - OUT: YUYV 320x260 @ 30fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:DemoArUco C++
   44 - OUT: YUYV 320x258 @ 60fps CAM: YUYV 320x240 @ 60fps MOD: JeVois:DemoARtoolkit C++
   45 - OUT: YUYV 320x256 @ 30fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:RoadNavigation C++
   46 - OUT: YUYV 320x254 @ 60fps CAM: YUYV 320x240 @ 60fps MOD: JeVois:ObjectTracker C++
   47 - OUT: YUYV 320x252 @ 30fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:ObjectDetect C++
   48 - OUT: YUYV 320x240 @ 60fps CAM: YUYV 320x240 @ 60fps MOD: JeVois:SaveVideo C++
   49 - OUT: YUYV 320x120 @ 30fps CAM: YUYV 160x120 @ 30fps MOD: JeVois:DemoBackgroundSubtract C++
   50 - OUT: YUYV 176x144 @ 120fps CAM: YUYV 176x144 @ 120fps MOD: JeVois:SaveVideo C++
   51 - OUT: YUYV 64x192 @ 25fps CAM: YUYV 320x240 @ 25fps MOD: JeVois:SalientRegions C++
   52 - OUT: GREY 640x480 @ 29fps CAM: YUYV 640x480 @ 29fps MOD: JeVois:EdgeDetection C++
   53 - OUT: GREY 640x480 @ 20fps CAM: YUYV 640x480 @ 20fps MOD: JeVois:PythonOpenCV Python
   54 - OUT: GREY 320x960 @ 45fps CAM: YUYV 320x240 @ 45fps MOD: JeVois:EdgeDetectionX4 C++
   55 - OUT: GREY 320x240 @ 59fps CAM: YUYV 320x240 @ 59fps MOD: JeVois:EdgeDetection C++
   56 - OUT: GREY 320x240 @ 30fps CAM: YUYV 320x240 @ 30fps MOD: JeVois:SuperPixelSeg C++
   57 - OUT: GREY 176x288 @ 100fps CAM: YUYV 176x144 @ 100fps MOD: JeVois:OpticalFlow C++
   58 - OUT: GREY 176x144 @ 120fps CAM: YUYV 176x144 @ 120fps MOD: JeVois:DemoEyeTracker C++
   59 - OUT: GREY 160x495 @ 60fps CAM: YUYV 160x120 @ 60fps MOD: JeVois:DemoCPUGPU C++
   60 - OUT: GREY 128x117 @ 5fps CAM: YUYV 160x120 @ 5fps MOD: JeVois:DenseSift C++

I executed jevois-daemon --videomapping=1, but it didn't work properly

Window running other modules is displayed and video displayed in window is stored.

help me, please

Answer 1

are you running two chained jevois-daemon instances, one in the camera and then one on your host?

your videomapping is ok. However, when you run jevois-daemon on host, you should disconnect your jevois camera and use a plain webcam that supports 640x480 instead. Otherwise, jevois will run one instance of jevois-daemon in the camera, and then your host will run a second instance on your host computer. If you don't want to see the demo with the baby running in your jevois camera, then you should add a video mapping on the SD card that configures for 640x480 as pass-through (dumb camera):

YUYV 640 480 30.0 YUYV 640 480 30.0 JeVois PassThrough

make sure that one is saved to your microSD card.

Please see here:

http://jevois.org/doc/Concepts.html

this section:

Note that usually one would either:

• Compile and run jevois-daemon in host mode on a host computer: In this case usually a dumb camera will be used (any regular webcam), all machine vision processing runs on the host CPU, and video output is to display; or
• Cross-compile jevois-daemon in platform mode and run it inside the JeVois smart camera: In this case the built-in camera sensor in the JeVois smart camera is used, all machine vision algorithms run on the small processor inside the JeVois smart camera, and video output is streamed over the USB link to a host. The host then runs a dumb video capture software.

But also note that chaining two jevois-daemon processes is possible, with one running inside the camera and a second one running on a host computer. For that to work, you need to make sure that the camera format requested by jevois-daemon running on the host exactly matches an available output format of the jevois-daemon running inside the smart camera. Such scenario may be useful to split image processing tasks between the JeVois smart camera and a host computer. For example, the JeVois smart camera could capture video and detect edges, and then transmit the edge maps over USB to a host computer that would capture them and process them further to detect and identify objects in the edge maps.