×

Help

General
Multi Shot is a HTML5 javascript-application for capturing the motion of a moving object in video by overlaying the different frames. The result is displayed in a single picture. This app can also be used in classrooms to teach students about motion and mechanics.

Getting started
Click on the menu icon at the right top corner and upload a video. Select the frames that you want to process and press "Process video".

Video formats
Video support depends on the browser and operating system. Almost all browsers support the common video containers (MP4, MOV, WebM and Ogg). On the other hand, be aware that support for the HEVC (H.265) codec is limited to mostly Safari. For instance, in case you want to analyse videos that have been recorded by an iPhone 7 or later, either change the format on your iPhone or convert the video using any online video converter. You can find the actual video format by clicking on the icon after uploading the video.

Video requirements
The camera should not move during the video.

Settings
You can choose to skip processing frames (faster). The frame rate is automatically obtained from the video meta data using the MediaInfo tool. In the event that this calculation stalls, you can set the frame rate manually. In most cases a frame rate of 30 or 60 frames per second (fps) is appropriate. A further explanation of the expert settings can be found in the source code.

Security/privacy
No data leaves your computer by using this website. All analysis is done locally on your own computer. This is a cookie-free website.

Webapp compatibility
Multi Shot does not work in Internet Explorer. In case you are still using IE, please consider moving to a modern browser like Chrome, Edge, Firefox or Safari.

Embed code
Teachers and others can embed Multi Shot in their own virtual learning environment (VLE). The embed code would be for example:

 <iframe src="https://jeroenvantilburg.nl/multishot/" width="100%" height="800"></iframe>

The width and height should be adjusted for your own environment.

Source code
The source code from this website is freely available on GitHub under the MIT license. The source code may be freely copied and hosted on every web server. Please make sure to keep the code up-to-date since new updates and bug fixes come out regularly.

Feedback and questions
To keep this website as user-friendly as possible I depend on your feedback. Please send your questions, improvements, bug reports, suggestions, comments and compliments to .

Acknowledgments
Many thanks to GitHub and GitHub Pages for hosting Multi Shot. Multi Shot depends on the following useful javascript libraries:

• jQuery: DOM manipulation
• Mediainfo.js: extract meta data from video files
• OpenCV: computer vision and machine learning software library

Other applications
Besides Multi Shot, there are a number of other, third-party applications that can be used to analyse videos. Select the application that suits your needs.

• Sony motion shot (free) is an app for Android and iOS that automatically renders the movement in a video into a single still image.

×

Settings

Basic
Scan rate:		Set the frequency at which the frames are scanned. A value of 1 means that each frames is processed.

Advanced
Frame rate:		Set the frame rate (frames per second) manually.
Video orientation:	0 90 180 270	Set the orientation of the video in degrees.
		Show expert settings.

Expert
Threshold:		Threshold to decide if it is well described by the background model or not. Does not influence the update of the background. Default is 16 (corresponds to 4 sigma).
Tg:		Threshold to decide when a sample is close to the existing components. If it is not close to any a new component will be generated. Smaller Tg leads to more generated components and higher Tg might make lead to small number of components but they can grow too large. Default is 9 (corresponds to 3 sigma).
Learning rate:		Contribution of each frame to the background model. Should be between 0 and 1. Negative value means that this contribution is automatically determined.
Number of mixtures:		The maximum allowed number of mixture components
Background ratio:		Threshold when the component becomes significant enough to be included into the background model. It is the TB=1-cf from the paper. Default is 0.5.
Shadow threshold:		Shadow is detected if the pixel is darker version of the background. This is a threshold on how much darker the shadow can be. Value of 0.5 (default) means that if pixel is more than 2 times darker then it is not shadow.
VarInit:		Initial variance for the newly generated components. Influences the speed of adaptation. A good guess should be made. Default is 400.
VarMin:		Minimal variance of the components. Influences the speed of adaptation. A good guess should be made. Default is 4.
VarMax:		Maximal variance of the components. Influences the speed of adaptation. A good guess should be made. Default is 1600
Complexity reduction:		Complexity reduction prior. Related to the number of samples needed to accept that a component actually exists. Default is 0.05.
Two pass:		In the first pass the background model is determined. In the second pass the (moving) foreground is selected. Unchecking this box means that only one pass is done in the processing.

×

Video File Info