Abstract

이동통신 채널과 같은 저대역 통신망에서는 비디오 전송을 위해서 H.263+와 같이 압축율이 높은 부호화 방법이 사용된다. 저대역폭을 통해 고압축 비디오 비트스트림을 전송할 경우, 패킷이 손실되면 영상의 품질이 심각하게 떨어진다. 본 논문에서는 H.263+ 부호화영상이 전송될 때 손상되는 움직임 벡터의 복원기법을 제안하였다. 본 논문에서는 손실된 블록의 움직임 벡터는 인접한 블록의 움직임 벡터와 높은 상관성을 갖는 다는 사실에 착안하여, 시공간적 오류은닉(Temporal-Spatial Error Concealment) 방법을 제안한다. 제안된 방법에서는 손실된 모션벡터를 인접한 블록의 움직임을 클러스터링하여 복원한다. 인접한 블록의 모션벡터는 ALA(Average Linkage Algorithm) 클러스터링 알고리즘에 따라 클러스터링되며, 각 클러스터의 대표값을 계산하여 후보 움직임 벡터 집합을 얻은 다음, 이들 후보의 움직임의 왜곡정도를 계산하여 왜곡이 최소인 움직임 벡터를 선택한다. 제안한 방법으로 복원한 영상의 화질에 대한 객관적, 주관적 평가에서 개선된 결과를 확인하였다. To transmit a video bit stream over low bandwith, such as mobile, channels, encoding algorithms for high bit rate like H.263+ are used. In transmitting video bit-streams, packet losses cause severe degradation in image quality. This paper proposes a new algorithm for the recovery of missing or erroneous motion vectors when H.263+ bit-stream is transmitted. Considering that the missing or erroneous motion vectors are closely related with those of neighboring blocks, this paper proposes a temporal-spatial error concealment algorithm. The proposed approach is that missing or erroneous Motion Vectors(MVs) are recovered by clustering the movements of neighboring blocks by their homogeneity. MVs of neighboring blocks we clustered according to ALA(Average Linkage Algorithm) clustering and a representative value for each cluster is determined to obtain the candidate MV set. By computing the distortion of the candidates, a MV with the minimum distortion is selected. Experimental results show that the proposed algorithm exhibits better performance in subjective and objective evaluation than existing methods.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.