Browse Prior Art Database

Adapting existing video file formats to enable tamper-proofing

IP.com Disclosure Number: IPCOM000011725D
Original Publication Date: 2003-Mar-12
Included in the Prior Art Database: 2003-Mar-12
Document File: 4 page(s) / 22K

Publishing Venue

Motorola

Related People

Paola Hobson: AUTHOR [+2]

Abstract

If we release digital video into the public domain, we may wish to create the video in such a way that it is impossible for anyone to alter it. Recent advances in video manipulation technology have been used to create systems that can remove or distort objects or people within each frame. In order to detect such attempts at tampering, the system described below embeds the video display information fields inside the video itself and removes the information from the normal (header) locations within the file. Any attempt to alter the video would result in corruption of the information stored within the frame. This would mean that when an attempt to view the video was made, there would be very noticeable visual effects.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 34% of the total text.

Adapting existing video file formats to enable tamper-proofing

by Paola Hobson and Jonathon Hare

Abstract

If we release digital video into the public domain, we may wish to create the video in such a way that it is impossible for anyone to alter it.� Recent advances in video manipulation technology have been used to create systems that can remove or distort objects or people within each frame.� � In order to detect such attempts at tampering, the system described below embeds the video display information fields inside the video itself and removes the information from the normal (header) locations within the file. Any attempt to alter the video would result in corruption of the information stored within the frame. This would mean that when an attempt to view the video was made, there would be very noticeable visual effects.

Introduction

This paper proposes modifications to existing video file formats such that the quality of the viewed video would be severely degraded if any attempt has been made to tamper with the contents of the video.� Releasing digital video into the public domain makes it susceptible to attacks which alter the content of the video, for example, in such a way as to cause misrepresentation.� Recent advances in video manipulation technology have been used to create systems that can process video and remove objects or people from each frame. It is easy to see how one person’s image could be removed and substituted with someone else’s image. This kind of video manipulation could cause great problems, for example in misrepresentation of news events, creation of fake scientific or public events, misrepresentation of criminal activities etc, so it is important that steps are taken to ensure that the video cannot be altered in this way.

Problem(s) To Be Solved

There are no known documented technologies that address this problem, although there are many known methods for applying a watermark to images and video.� Watermarking systems can be used to validate authentication and ownership of video sequences.� They can also be used to detect if tampering of the video has occurred.� However, such systems do not prevent the video being viewed in a pirated or tampered state.� If an attacker or pirate believes the watermark will not be tested for until well after they have sold on the video material (or otherwise distributed it), they will not hesitate to tamper or copy it.

Proposed Solution to the Problem(s)

The system described in this paper creates a file which will effectively destroy itself if it is tampered with.� It works by embedding the video presentation information fields inside the video itself and removing the information from the normal locations (for example header) within the file.

For example the file format may normally specify the size information of the frame in a certain field and this could be removed from the file header (or time-to-sample atom in the case of QuickTime) and embedded within the video pixels of th...