Wikitude

Getting wikitude to recognise images

Hi,

I have my wtc file generated and 3 test images which should fire 3 videos. The imagerecognition.js file is set like this

createOverlays: function createOverlaysFn() {
        /*
            First an AR.Tracker needs to be created in order to start the recognition engine. It is initialized with a URL specific to the target collection. Optional parameters are passed as object in the last argument. In this case a callback function for the onLoaded trigger is set. Once the tracker is fully loaded the function worldLoaded() is called.

            Important: If you replace the tracker file with your own, make sure to change the target name accordingly.
            Use a specific target name to respond only to a certain target or use a wildcard to respond to any or a certain group of targets.
        */
        this.tracker = new AR.ClientTracker("assets/tracker.wtc", {
            onLoaded: this.worldLoaded
        });

        /*
            The next step is to create the augmentation. In this example an image resource is created and passed to the AR.ImageDrawable. A drawable is a visual component that can be connected to an IR target (AR.Trackable2DObject) or a geolocated object (AR.GeoObject). The AR.ImageDrawable is initialized by the image and its size. Optional parameters allow for position it relative to the recognized target.
        */

        /* Create overlay for page one */
        var imgOne = new AR.ImageResource("assets/FG7809.png");
        var overlayOne = new AR.ImageDrawable(imgOne, 1, {
            offsetX: 0,
            offsetY: 0
        });

        var video = new AR.VideoDrawable("assets/wax2012.mp4", .5, {
    offsetX: 0,
    offsetY: 0,
    onLoaded: function videoLoaded() {
    video.enabled = true;
    },
    onPlaybackStarted: function videoPlaying () {
    video.playing = true;
        video.enabled = true;
    },
    onFinishedPlaying: function videoFinished () {
        video.playing = false;
        video.enabled = false;
    },
    onClick: function videoClicked () {
        if (video.playing) {
            video.pause();
            video.playing = false;
        } else {
            video.resume();
            video.playing = true;
        }
    }
});

var pageOne = new AR.Trackable2DObject(this.tracker, "imgOne", {
            drawables: {
                cam:
            },
            onEnterFieldOfVision: function onEnterFieldOfVisionFn () {
        if (video.playing) {
            video.pause();
        }
    },
    onExitFieldOfVision: function onExitFieldOfVisionFn () {
        if (video.playing) {
            video.pause();
        }
    }
    });

If I change the imgOne in the pageOne definition to * and add in display.alert("hello"); the alert shows. However, no video.

I have twenty or thirty trigger images with the same number of videos to display, so I really need to make sure that I have the imagerecognition.js file set up correctly.

Am I doing this correctly and what should the parameter inside of the quotes on pageOne = (this,tracker, "...") be? I have a project where it is imgOne and another where it is pageOne.

Thanks

PFJ

Hi there!

Please try remote WebView debugging to figure out potential JS errors and apply your changes step-by-step from the provided sample applications.

best regards

Hi,

The web console is showing no errors at all which is really annoying! If I put the trigger on pageOne to "*", the it looks to be working, so the only thing I can summise is that what I have for * is incorrect in my other pages.

My full source can be seen here http://pastebin.com/Py3TG5ne

Any comments would be appreciated

PFJ

Hi Paul!

In case you just have 3 targetImages / videos you may also just create the trackables in a loop using same approach as in the sample application's "multiple targets".

Using "*" as targetName is fine if you use the very same augmentations but needs some extra work (smart caching and add/removeCamdrawables) in terms of asset management, which is something to consider for more than 5 videos.

Best regards,
Andreas