Image Capture is an API to capture still images and configure camera hardware settings. This API is available in Chrome 59 on Android and desktop. We've also published an ImageCapture polyfill library.
The API enables control over camera features such as zoom, brightness, contrast, ISO and white balance. Best of all, Image Capture allows you to access the full resolution capabilities of any available device camera or webcam. Previous techniques for taking photos on the Web have used video snapshots, which are lower resolution than that available for still images.
An ImageCapture
object is constructed with a MediaStreamTrack
as source. The
API has then two capture methods takePhoto()
and grabFrame()
and ways to
retrieve the capabilities and settings of the camera, and to change those
settings.
Construction
The Image Capture API gets access to a camera via a MediaStreamTrack
obtained
from getUserMedia()
:
navigator.mediaDevices.getUserMedia({video: true})
.then(gotMedia)
.catch(error => console.error('getUserMedia() error:', error));
function gotMedia(mediaStream) {
const mediaStreamTrack = mediaStream.getVideoTracks()[0];
const imageCapture = new ImageCapture(mediaStreamTrack);
console.log(imageCapture);
}
You can try out this code from the DevTools console.
Capture
Capture can be done in two ways: full frame and quick snapshot. takePhoto()
returns a Blob
, the result of a single photographic exposure,
which can be downloaded, stored by the browser or displayed in an <img>
element. This method uses the highest available photographic camera resolution.
For example:
const img = document.querySelector('img');
// ...
imageCapture.takePhoto()
.then(blob => {
img.src = URL.createObjectURL(blob);
img.onload = () => { URL.revokeObjectURL(this.src); }
})
.catch(error => console.error('takePhoto() error:', error));
grabFrame()
returns an ImageBitmap
object, a snapshot of live video,
which could (for example) be drawn on a <canvas
> and then post-processed to
selectively change color values. Note that the ImageBitmap
will only have the
resolution of the video source — which will usually be lower than the camera's
still-image capabilities. For example:
const canvas = document.querySelector('canvas');
// ...
imageCapture.grabFrame()
.then(imageBitmap => {
canvas.width = imageBitmap.width;
canvas.height = imageBitmap.height;
canvas.getContext('2d').drawImage(imageBitmap, 0, 0);
})
.catch(error => console.error('grabFrame() error:', error));
Capabilities and settings
There are a number of ways to manipulate the capture settings, depending on
whether the changes would be reflected in the MediaStreamTrack
or can only be
seen after takePhoto()
. For example, a change in zoom
level is immediately
propagated to the MediaStreamTrack
whereas the red eye reduction, when set, is
only applied when the photo is being taken.
"Live" camera capabilities and settings are manipulated via the preview
MediaStreamTrack
: MediaStreamTrack.getCapabilities()
returns a
MediaTrackCapabilities
dictionary with the concrete supported capabilities and the ranges or allowed
values, e.g. supported zoom range or allowed white balance modes.
Correspondingly, MediaStreamTrack.getSettings()
returns a
MediaTrackSettings
with the concrete current settings. Zoom, brightness, and torch mode belong to
this category, for example:
var zoomSlider = document.querySelector('input[type=range]');
// ...
const capabilities = mediaStreamTrack.getCapabilities();
const settings = mediaStreamTrack.getSettings();
if (capabilities.zoom) {
zoomSlider.min = capabilities.zoom.min;
zoomSlider.max = capabilities.zoom.max;
zoomSlider.step = capabilities.zoom.step;
zoomSlider.value = settings.zoom;
}
"Non-Live" camera capabilities and settings are manipulated via the
ImageCapture
object: ImageCapture.getPhotoCapabilities()
returns a
PhotoCapabilities
object that provides access to "Non-Live" available camera capabilities.
Correspondingly, starting in Chrome 61, ImageCapture.getPhotoSettings()
returns a
PhotoSettings
object with the concrete current settings. The photo resolution, red eye
reduction and flash mode (except torch) belong to this section, for example:
var widthSlider = document.querySelector('input[type=range]');
// ...
imageCapture.getPhotoCapabilities()
.then(function(photoCapabilities) {
widthSlider.min = photoCapabilities.imageWidth.min;
widthSlider.max = photoCapabilities.imageWidth.max;
widthSlider.step = photoCapabilities.imageWidth.step;
return imageCapture.getPhotoSettings();
})
.then(function(photoSettings) {
widthSlider.value = photoSettings.imageWidth;
})
.catch(error => console.error('Error getting camera capabilities and settings:', error));
Configuring
"Live" camera settings can be configured via the preview
MediaStreamTrack
's applyConstraints()
constraints
, for example:
var zoomSlider = document.querySelector('input[type=range]');
mediaStreamTrack.applyConstraints({ advanced: [{ zoom: zoomSlider.value }]})
.catch(error => console.error('Uh, oh, applyConstraints() error:', error));
"Non-Live" camera settings are configured with the takePhoto()
's optional
PhotoSettings
dictionary, for example:
var widthSlider = document.querySelector('input[type=range]');
imageCapture.takePhoto({ imageWidth : widthSlider.value })
.then(blob => {
img.src = URL.createObjectURL(blob);
img.onload = () => { URL.revokeObjectURL(this.src); }
})
.catch(error => console.error('Uh, oh, takePhoto() error:', error));
Camera capabilities
If you run the code above, you'll notice a difference in dimensions between the
grabFrame()
and takePhoto()
results.
The takePhoto()
method gives access to the camera's maximum resolution.
grabFrame()
just takes the next available VideoFrame
in the MediaStreamTrack
inside the renderer process, whereas takePhoto()
interrupts the MediaStream
,
reconfigures the camera, takes the photo (usually in a compressed format,
hence the Blob
) and then resumes the MediaStreamTrack
. In essence, this means
that takePhoto()
gives access to the full still-image resolution
capabilities of the camera. Previously, it was only possible to 'take a photo' by
calling drawImage()
on a canvas
element, using a video as the source (as per the
example here).
More information can be found in the README.md section.
In this demo, the <canvas>
dimensions are set to the resolution of the video
stream, whereas the natural size of the <img>
is the maximum still-image
resolution of the camera. CSS, of course, is used to set the display
size of both.
The full range of available camera resolutions for still images can be get and set
using the MediaSettingsRange
values for PhotoCapabilities.imageHeight
and
imageWidth
. Note that the minimum and maximum width and height constraints for
getUserMedia()
are for video, which (as discussed) may be different from the
camera capabilities for still images. In other words, you may not be able to
access the full resolution capabilities of your device when saving from
getUserMedia()
to a canvas. The WebRTC resolution constraint demo
shows how to set getUserMedia()
constraints for resolution.
Anything else?
The Shape Detection API works well with Image Capture:
grabFrame()
can be called repeatedly to feedImageBitmap
s to aFaceDetector
orBarcodeDetector
. Find out more about the API from Paul Kinlan's blog post.The Camera flash (device light) can be accessed via
FillLightMode
inPhotoCapabilities
, but the Torch mode (flash constantly on) can be found in theMediaTrackCapabilities
.
Demos and code samples
Support
- Chrome 59 on Android and desktop.
- Chrome Canary on Android and desktop previous to 59 with Experimental Web Platform features enabled.