Video and Audio Annotations – New Functionality in OpenAnnotate

TSG has just finished our latest addition to OpenAnnotate to support annotations of various file formats including mp4 (video) and wav/mp3/aac (audio).  This post will present our approach as well as a short video demonstration.

Video and Audio Annotations – Adding annotations but protecting the integrity of the video/audio file

One of the big advantages of PDF Annotations has always been the ability of PDFs to support a separate layer outside of the PDF file itself.  For OpenAnnotate, we leverage the Adobe XFDF standard.  OpenAnnotate, as well as other PDF viewing products, can display the PDF as well as one or multiple layers of XFDF on top of the PDF without affecting the integrity of the PDF file.

Video files present a slightly different problem as:

• There are multiple video formats mostly focused on just playing the video without a standard for adding annotations.
• Many of our clients view the video itself (for example security footage) as a piece of evidence. Any annotations cannot manipulate the video and likely serve as a pointer to certain times on the video.
• While video players allow viewing of the video, placing the annotation on the video image can be difficult to find and require deciding on how long the annotation should appear.

Audio files present even more problems, since there is no way to visually represent comments/bookmarks into an audio file directly.

OpenAnnotate addresses the issues above by annotating on a separate PDF file with a timeline (and potentially a written transcript) that integrates in with the video/audio player.  In this manner:

• Annotations can quickly be viewed and selected to view the specific point in the video, serving as a kind of “bookmark” to various points of the video/audio file that are of interest with comments.
• Video/audio player can communicate with the PDF annotations to scroll through annotations as the video/audio is playing.
• OpenAnnotate relies on an underlying ECM repository (Alfresco, Documentum or Hadoop) to keep the relationship between the video/audio file and the annotations.
• The PDF rendition is a separate record that can stand on its own in the case of audio files that have no physical place to represent comments/bookmarks.

Below is a quick demo of the solution. Let us know your thoughts in the comments below.