# AnySurface API DRAFT

## glossary:

things to need to name:

* what is currently refered to as a `match`, this is an important
  object in the system and deserves a good name.


## Configuration

description of what configuration options are necessary and important.

* html target
* imageLoader
* Assignment Sections


TODO: fill out configuration object

``` javascript
const config = {
    canvasID: "canvas",
}

const mySurface = new AnySurface(config);
```


## Initialization
how to instantiate and start interacting with anySurface objcts.

initialization process description:

* Camera Selection
* Camera Alignment
* Calibrate
  * background
* Assignment


Event driven callback registration system of some kind...

```
mySurface.on(AnySurface.CalibrateComplete, () => {
    // do the work
});


mySurface.on(AnySurface.init, () => {
    model.start();
});

mySurface.init();
```


## Assignment
Generation of surface elements that will act as computational objects or triggers.

user creates sections, that deliniates a 'training zone' for object recognition.

```
mySurface.setSections(sections);
mySurface.previewAssignment(); -> cause projection to show assignment grid
```



## Detection
Returns a list of 'matches' detected by anySurface.

description of data struction returned
match:



```javascript
{
    imageBlob:      {...}
    trainingBlob:   {
        optionalQrCode,
        image(RBG),
        mask,

    }
    position:       {...}   // cords for both projector space and canvas space
    QR:             {...}   // meta data about object
}
```


```javascript

mySurface.detect(elements => {
    // interact with matches to perhaps update model
});
```



### Monitoring
basic event notification for when anySurface detects the objects in the projection have changed.
this runs the detection routine.



## Positioning
api for mapping between camera coords to screen coords


```javascript
const screenXY = mySurface.getCanvasPositionFromCamera(camX, camY);
```

api for mapping between screen coords and camera?