The fast-image-sequence is a powerful package that allows you to display a sequence of images at a high frame rate on your website. It can be used to create smooth animations, 360 product views or video-like sequences from a series of images. Zero dependencies.
The FastImageSequence supports various options for customizing the behaviour of the image sequence, such as preloading all images, using a worker to handle tar files, and more.
This is a build from the repository's example/ directory.
Add @mediamonks/fast-image-sequence
to your project:
npm i @mediamonks/fast-image-sequence
Creating a FastImageSequence instance and playing an image sequence.
import {FastImageSequence} from '@mediamonks/fast-image-sequence';
const options = {
frames: 100,
src: {
imageURL: (index) => `path/to/your/image/sequence/image${index}.jpg`,
},
loop: true,
objectFit: 'cover',
};
const sequence = new FastImageSequence(containerElement, options);
sequence.play();
In the options object, you must set the numbers of frames
and an imageURL
. The imageURL
is a function that
takes an index as a parameter and returns a string representing the URL of the image at that index in the sequence.
If you want, you can load images from a tar file. This can be useful when you want to preload all images at once with
minimal memory usage. You can use a tar file by setting the tarURL
option. You also need to set the imageURL
, which
in this case returns the image's URL in the tar file.
const options = {
frames: 100,
src: {
tarURL: 'path/to/your/tar/file.tar',
imageURL: (index) => `image${index}.jpg`,
},
loop: true,
objectFit: 'cover',
};
You can also set multiple sources for the FastImageSequence class. This can be useful when you want to load images from different sources, such as a tar file and an image URL. You can set an array of ImageSource in the src option. Each ImageSource can have its imageURL and tarURL.
The FastImageSequence class will try to load images from the first source in the array. If an image is not available yet, it will try to load it from the next source in the array, etc. Finally, the best matching available image will be rendered.
import {FastImageSequence} from '@mediamonks/fast-image-sequence';
const options = {
frames: 100,
src: [
{
// First try to display a highres image from an image URL
imageURL: (index) => `path/to/your/image/sequence/highres_image${index}.jpg`,
maxCachedImages: 16,
},
{
// Fallback if highres image is not loaded yet: serve a low res image from the tar file
tarURL: 'path/to/your/tar/file.tar',
imageURL: (index) => `lowres_image${index}.jpg`,
maxCachedImages: 32,
},
],
loop: false,
objectFit: 'contain',
};
const sequence = new FastImageSequence(containerElement, options);
sequence.play();
The example above can be useful when you have a large image sequence with high-resolution images or if the user randomly jumps to any frame in the sequence and you want a fast response time.
Note that you can store a subset of the images (all even images, for example) in the tar file. The FastImageSequence will automatically fall back to the best matching available image when rendering a frame.
-
frames:
number
- Number of frames in the sequence. Required. -
src:
ImageSourceOptions[] | ImageSourceOptions
The source of the images for the FastImageSequence class. It can either be an array of ImageSourceOptions or a single ImageSourceOptions instance. -
loop:
boolean
- Whether the sequence should loop. Default:false
-
objectFit:
'contain' | 'cover'
- How the image should fit the canvas. Default:'cover'
-
horizontalAlign:
number
- Horizontal image alignment. Default:0.5
-
verticalAlign:
number
- Vertical image alignment. Default:0.5
-
poster:
string | undefined
- URL of the poster image. Optional. -
fillStyle:
string
- Fill style of the canvas. Default:'#00000000'
-
clearCanvas:
boolean
- Clear the canvas before drawing. Default:false
-
showDebugInfo:
boolean
- Show debug info. Default:false
-
name:
string
- Name of the FastImageSequence instance. Default:'FastImageSequence'
- imageURL:
(index: number) => string
- Callback returning the URL of an image given its index. - tarURL:
string | undefined
- URL of the tar file containing images. Optional. - useWorker:
boolean
- Use a worker to fetch images. Default:!isMobile()
- maxConnectionLimit:
number
- Maximum concurrent connections for fetching images. Default:4
- maxCachedImages:
number
- Number of images to cache. Default:32
- available:
((index: number) => boolean) | undefined
- Callback returning whether an image is available. Optional. - image:
((index: number) => Promise<CanvasImageSource>) | undefined
- Callback returning the image given its index. Optional. - timeout:
number
- Only start loading an image if the same frame is visible for this time (in milliseconds). Optional.
The following methods are available for a FastImageSequence object:
This method registers a tick function to be called on every frame update.
sequence.tick(func);
func
is a function that will be called on every frame update.
This method starts playing the image sequence at a specified frame rate.
sequence.play(fps);
fps
is an optional parameter determining the frame rate at which to play the sequence. If not provided, it defaults to30
. Note that the fps can also be negative, which will play the sequence in reverse.
This method stops playing the image sequence.
sequence.stop();
This getter method returns a boolean indicating whether the image sequence is playing.
const playingStatus = sequence.playing;
This getter method returns a boolean indicating whether the image sequence is paused.
const pausedStatus = sequence.paused;
This getter-and-setter method retrieves or sets the image sequence's current progress (0-1).
// Get the progress
const currentProgress = sequence.progress;
// Set the progress
sequence.progress = value;
value
is a number that sets the current progress of the image sequence.
This is a getter method that retrieves the current load progress of the image sequence.
const currentLoadProgress = sequence.loadProgress;
The loadProgress
property returns a number between 0 and 1, representing the current load progress of the image
sequence. Note that the value can decrease if the sequence is played and new frames are loaded.
This method gets the image of a specific frame and returns a Promise that resolves with the image of the frame.
sequence.getFrameImage(index);
index
is the index of the frame.
Register a callback function that is called with the progress of the loading. The function returns a promise that resolves when progress reaches 1.
sequence.onLoadProgress(onProgress);
sources[index:number].setMaxCachedImages(maxCache: number, onProgress?: (progress: number) => void): Promise
This method sets the number of images that should be preloaded and cached in memory for an input source and returns a Promise that resolves when all these images are preloaded and cached.
sequence.sources[0].setNumberOfCachedImages(maxCache, onProgress);
maxCache
is the number of images to cache. This should be a positive integer.onProgress
is an optional callback function whenever the loading progress changes. It receives the current progress as a number between 0 and 1.
The returned Promise resolves when loadProgress reaches 1.
A tarball is a collection of files and directories stored in a single file. In the context of FastImageSequence
, you
can use a tarball to store (low-resolution preview) images for your image sequence. This can be particularly useful when
you want to seek or jump in the sequence quickly or when the internet speed is low.
To create a tarball with images, you can follow these steps:
-
Prepare your images: Make sure all your images are in a single directory. The images should be in a sequence and named consistently (for example,
image1.jpg
,image2.jpg
,image3.jpg
, etc.). -
Create a tarball file. I have created an easy-to-use online tool for this: Tar File Creator (no ads, login, etc). Drag and drop your selection of images onto the page, and a tarball will be generated that you can download.
-
Alternatively, use a tar tool to create the tarball: if you are comfortable with the command line, you can use the
tar
command in Unix-based systems like this:cd /path/to/your/images tar -cvf sequence.tar -C ./ .
Once you have your tarball, you can use it with FastImageSequence
by setting the tarURL
option to the URL of your
tarball and implement the imageURL
to return the URL of an image in the tarball, given its index.
I created an easy-to-use online tool for this: Tar File Creator. Drag and drop your selection of images onto the page, and a tar file will be generated that you can download. You can also use a tar tool to create the tar file yourself.
You can set the maxCachedImages
option to 8. The FastImageSequence will only preload and cache the first 8 images. You can then set the maxCachedImages
option to a higher number to preload the rest of the images later.
See this example for more information.
I have an image sequence of low-res images and want to download a high-res image when the user stops at a frame. How can I do this?
You can set multiple sources using the src
option. The FastImageSequence will try to load images from the first source in the array. If an image is not available yet, it will try to load it from the next source in the array, etc. Finally, the best matching available image will be rendered.
By setting a timeout
option, you can control when the FastImageSequence should start loading an image. For example:
const options = {
frames: 100,
src: [
{
// First try to display a highres image from an image URL
imageURL: (index) => `path/to/your/image/sequence/highres_image${index}.jpg`,
maxCachedImages: 1,
timeout: 16, // only start loading an image if the same frame is visible for 16ms
},
{
// Default: serve a low res image from the sequence
imageURL: (index) => `path/to/your/image/sequence/lowres_image${index}.jpg`,
maxCachedImages: 32,
},
],
loop: false,
objectFit: 'contain',
};
const sequence = new FastImageSequence(containerElement, options);
Yes, you can download the tar file yourself and create a data URL from it. You can then use this data URL as the tarURL
option.
See this example for more information.
To build fast-image-sequence, ensure that you have Git and Node.js installed.
Clone a copy of the repo:
git clone https://github.com/mediamonks/fast-image-sequence-renderer.git
Change to the fast-image-sequence-renderer directory:
cd fast-image-sequence-renderer
Install dev dependencies:
npm i
Build package:
npm run build