Browse Prior Art Database

MPEG-4 Error Concealment Using Motion Vector Data Recovery

IP.com Disclosure Number: IPCOM000024000D
Publication Date: 2004-Apr-01
Document File: 5 page(s) / 2M

Publishing Venue

The IP.com Prior Art Database

Abstract

In concealing video packets with corrupted motion partition, the decoder attempts to utilize uncorrupted received motion information. Image smoothness is evaluated for varying amount of decoded motion information. The amount of utilized decoded motion information is selected to maximize image smoothness.

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 52% of the total text.

         Sony Corporation

         Sony Electronics Inc.

         IPD Case #50R4720

Title:

MPEG-4 Error Concealment Using Motion Vector Data Recovery

Abstract:

In concealing video packets with corrupted motion partition, the decoder attempts to utilize uncorrupted received motion information. Image smoothness is evaluated for varying amount of decoded motion information. The amount of utilized decoded motion information is selected to maximize image smoothness.

Inventor:

Joseph C Chan

Invention Disclosure Data Sheet:

A. Existing problems

In an MPEG-4 decoder used for real-time streaming, when an error is detected in the motion partition of a video packet, all decoded information of that packet is usually discarded. The region of the video frame corresponding to this corrupted packet is then generated by error concealment process. In situations without scene changes, motion compensated temporal replacement is an effective error concealment technique as long as motion information without much error can be obtained by the decoder.

One approach in obtaining the motion information is to use average or median of motion vectors from neighboring uncorrupted macroblocks. This technique is simple but sometimes yields inadequate performance when the actual motion vector of neighboring macroblocks differs. Another approach involves use of motion estimation in the decoder (as in IPD 50P4109). This generally yields better visual results but requires significant processing.

B. Invention Description

In concealing video packets with corrupted motion partition, the decoder attempts to utilize uncorrupted received motion information. Image smoothness is evaluated for varying amount of utilized decoded motion information. The amount of decoded motion information is selected to maximize image smoothness. The flow chart of this process is shown in Figure I.

The numbering of the following sections are descriptions of the steps in Figure 1 with same numbering.

(1) Packet Error Detection

Errors of a video packet can be detected if invalid variable length code (VLC), inconsistent resynchronization header information or out-of-range motion vectors are received. Note that since variable length codes are used, the location where the decoder detects an error (i.e. where invalid VLC occur) is usually some undetermined number of bits away from where the actual start of the error. Furthermore, since motion information is also differentially encoded, error in decoded motion information may persist after the location of bit error. Figure 2 shows an example of a corrupted video packet. The techniques for packet error detection are outside the scope of this invention.

(2) Initialization:

In MPEG, a video frame is divided into macroblocks. Each video packet contains information for consecutive macroblocks from left to right and then top to bottom. The number of macroblocks (N) represented by each video packet can be determined by the change in macroblock number field of the video packet header.

In this ste...