3d-view

This module is a generic interface which synchronizes several existing view interactions

• turntable-camera-controller
• orbit-camera-controller
• matrix-camera-controller

Each camera controller proceeds by appending events onto a log. The current camera state is obtained by interpolating between these events.

API

Try a more complete demo here

var now            = require('right-now')
var bunny          = require('bunny')
var perspective    = require('gl-mat4/perspective')
var fit            = require('canvas-fit')
var createContext  = require('gl-context')
var createAxes     = require('gl-axes')
var createMesh     = require('gl-simplicial-complex')
var createCamera   = require('3d-view')

//Set up WebGL
var canvas = document.createElement('canvas')
document.body.appendChild(canvas)
window.addEventListener('resize', fit(canvas), false)
var gl = createContext(canvas, {}, render)

//Create objects for rendering
var bounds = [[-10,-10,-10], [10,10,10]]
var mesh = createMesh(gl, {
    cells: bunny.cells,
    positions: bunny.positions,
    colormap: 'jet'
  })
var axes = createAxes(gl, {
    bounds: bounds,
    tickSpacing: [1,1,1],
    textSize: 0.05
  })

//Set up camera
var projectionMatrix = new Array(16)
var camera = createCamera({
  center:  [
    0.5*(bounds[0][0]+bounds[1][0]),
    0.5*(bounds[0][1]+bounds[1][1]),
    0.5*(bounds[0][2]+bounds[1][2]) ],
  eye: [0, 0, bounds[1][2]],
  distanceLimits: [1, 1000]
})

//Create mode drop down
var modeSelect = document.createElement('select')
camera.modes.forEach(function(mode) {
  modeSelect.add(new Option(mode, mode))
})
modeSelect.style.position = 'absolute'
modeSelect.style.left = '10px'
modeSelect.style.top = '10px'
modeSelect.style['z-index'] = 10
document.body.appendChild(modeSelect)


//Hook event listeners
var lastX = 0, lastY = 0

document.oncontextmenu = function(e) { 
  e.preventDefault()
  e.stopPropagation()
  return false 
}

modeSelect.addEventListener('change', function(ev) {
  camera.setMode(modeSelect.value)
})

canvas.addEventListener('mousemove', function(ev) {
  var dx =  (ev.clientX - lastX) / gl.drawingBufferWidth
  var dy = -(ev.clientY - lastY) / gl.drawingBufferHeight
  if(ev.which === 1) {
    if(ev.shiftKey) {
      //zoom
      camera.rotate(now(), 0, 0, dx)
    } else {
      //rotate
      camera.rotate(now(), dx, dy)
    }
  } else if(ev.which === 3) {
    //pan
    camera.pan(now(), dx, dy)
  }
  lastX = ev.clientX
  lastY = ev.clientY
})

canvas.addEventListener('wheel', function(e) {
  camera.pan(now(), 0, 0, e.deltaY)
})

//Redraw frame
function render() {

  //Update camera parameters
  var t = now()
  camera.idle(t - 20)
  camera.flush(t - 100)
  camera.recalcMatrix(t-25)
  
  //Compute parameters
  var cameraParams = {
    view: camera.computedMatrix,
    projection: perspective(
      [],
      Math.PI/4.0,
      gl.drawingBufferWidth/gl.drawingBufferHeight,
      0.1,
      1000.0)
  }

  //Draw everything
  gl.viewport(0, 0, gl.drawingBufferWidth, gl.drawingBufferHeight)
  gl.enable(gl.DEPTH_TEST)
  axes.draw(cameraParams)
  mesh.draw(cameraParams)
}

Constructor

var camera = require('3d-view')([options])

Methods

camera.idle(t)

Idles the camera at time t

• t is the time to idle at

camera.flush(t)

Flush all events in camera state before time t

• t is the cut off time for the flush

camera.modes

An array of modes supported by the camera

camera.setMode(mode)

Sets the camera mode

• mode is the new mode. Must be either turntable, orbit or matrix

camera.getMode()

Retrieves the current camera mode

camera.lookAt(t, eye, center, up)

Reset camera position to focus on a specified target

• t is the time of the event
• eye is the position of the camera
• center is the target of the camera
• up is a vector pointing up

camera.rotate(t, yaw, pitch, roll)

Rotates the camera incrementally by some amount

• t is the time of the input event
• yaw is the amount to rotate by along y-axis in radians
• pitch is the amount to rotate by along x-axis in radians
• roll is the amount to rotate by along z-axis in radians

camera.pan(t, dx, dy, dz)

Pans the camera in local (view relative) coordinates

• t is the time of the event
• dx,dy,dz is the amount to move

camera.translate(t, dx, dy, dz)

Translates the camera in world (absolute global) coordinates

• t is the time of the event
• dx,dy,dz is the amount to move

camera.setMatrix(t, matrix)

Sets the camera matrix to some fixed 4x4 matrix

• t is the time of the event
• matrix is the new camera matrix

camera.setDistance(t, r)

Sets camera distance at time t

• t is the time of the event
• r is the new camera distance

camera.setDistanceLimits(lo, hi)

Sets bounds on the camera distance

camera.getDistanceLimits([out])

Retrieves the camera limits

camera.recalcMatrix(t)

Recomputes all matrix properties at time t

camera.computedMatrix

The computed 4x4 matrix of the camera

camera.computedEye

The computed 3d eye vector for the camera

camera.computedUp

Computed up vector (initialized when calling recalcMatrix)

camera.computedCenter

Computed camera center point

camera.computedRadius

Computed log(radius)

License

(c) 2015 Mikola Lysenko. MIT License