TY - GEN
T1 - An efficient frame loss error concealment scheme based on tentative projection for H.264/AVC
AU - Sun, Hao
AU - Liu, Peilin
AU - Wang, Jun
AU - Goto, Satoshi
PY - 2010
Y1 - 2010
N2 - In this paper, we proposed an efficient frame loss error concealment scheme based on reverse projection. Frame loss often happens in low bit-rate video transmissions. This method utilized bi-direction motion vector extrapolation to generate a coverage motion vector candidate list. Then each motion vector in the candidate list is tentatively projected into the previous frame and the next frame. A distortion estimation is used to choose a best MV from the previous and the next frame each. A final MV is obtained with a liner interpolation of the chosen MVs. Experimental results have demonstrated that the proposed method improves the objective performance in terms of PSNR as well as subjective visual quality and the tentative projection is efficient.
AB - In this paper, we proposed an efficient frame loss error concealment scheme based on reverse projection. Frame loss often happens in low bit-rate video transmissions. This method utilized bi-direction motion vector extrapolation to generate a coverage motion vector candidate list. Then each motion vector in the candidate list is tentatively projected into the previous frame and the next frame. A distortion estimation is used to choose a best MV from the previous and the next frame each. A final MV is obtained with a liner interpolation of the chosen MVs. Experimental results have demonstrated that the proposed method improves the objective performance in terms of PSNR as well as subjective visual quality and the tentative projection is efficient.
KW - Error Concealment
KW - Frame Loss
KW - H.264/AVC
KW - Tentative Projection
UR - http://www.scopus.com/inward/record.url?scp=78049436212&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=78049436212&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-15696-0_37
DO - 10.1007/978-3-642-15696-0_37
M3 - Conference contribution
AN - SCOPUS:78049436212
SN - 3642156959
SN - 9783642156953
VL - 6298 LNCS
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 394
EP - 404
BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
T2 - 11th Pacific Rim Conference on Multimedia, PCM 2010
Y2 - 21 September 2010 through 24 September 2010
ER -