Quality adaptive trained filters for compression artifacts removal

L. Shao; J. Wang; I. Kirenko; G. Haan, de

Quality adaptive trained filters for compression artifacts removal

L. Shao
, J. Wang
, I. Kirenko
, G. Haan, de

Electronic Systems

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

191 Downloads (Pure)

Samenvatting

A compression artifacts removal algorithm that is adaptive to the artifact visibility level of the input video signal is proposed. The artifact visibility is determined per frame by the ratio of the accumulated gradient on the block edges to that of the remaining area. The filtering of each video frame is optimized using a least mean square mechanism which trains on pairs of target images and decompressed images of similar quality as the input frame. Experimental results show that the proposed approach outperforms several recent methods in coding artifact reduction.

Originele taal-2	Engels
Titel	Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference, March 31 2008-April 4 2008
Pagina's	89-900
Status	Gepubliceerd - 2008
Evenement	2008 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2008) - Las Vegas, Verenigde Staten van Amerika Duur: 30 mrt. 2008 → 4 apr. 2008

Congres

Congres	2008 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2008)
Verkorte titel	ICASSP 2008
Land/Regio	Verenigde Staten van Amerika
Stad	Las Vegas
Periode	30/03/08 → 4/04/08
Ander	ICASSP 2008

Toegang tot document

Metis220723Definitieve gepubliceerde versie, 570 KB

Citeer dit

@inproceedings{296742c492834b598329e2eb44a5df2f,

title = "Quality adaptive trained filters for compression artifacts removal",

abstract = "A compression artifacts removal algorithm that is adaptive to the artifact visibility level of the input video signal is proposed. The artifact visibility is determined per frame by the ratio of the accumulated gradient on the block edges to that of the remaining area. The filtering of each video frame is optimized using a least mean square mechanism which trains on pairs of target images and decompressed images of similar quality as the input frame. Experimental results show that the proposed approach outperforms several recent methods in coding artifact reduction.",

author = "L. Shao and J. Wang and I. Kirenko and \{Haan, de\}, G.",

year = "2008",

language = "English",

pages = "89--900",

booktitle = "Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference, March 31 2008-April 4 2008",

note = "2008 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2008), ICASSP 2008 ; Conference date: 30-03-2008 Through 04-04-2008",

}

Shao, L, Wang, J, Kirenko, I & Haan, de, G 2008, Quality adaptive trained filters for compression artifacts removal. in Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference, March 31 2008-April 4 2008. blz. 89-900, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2008), Las Vegas, Nevada, Verenigde Staten van Amerika, 30/03/08.

TY - GEN

T1 - Quality adaptive trained filters for compression artifacts removal

AU - Shao, L.

AU - Wang, J.

AU - Kirenko, I.

AU - Haan, de, G.

PY - 2008

Y1 - 2008

N2 - A compression artifacts removal algorithm that is adaptive to the artifact visibility level of the input video signal is proposed. The artifact visibility is determined per frame by the ratio of the accumulated gradient on the block edges to that of the remaining area. The filtering of each video frame is optimized using a least mean square mechanism which trains on pairs of target images and decompressed images of similar quality as the input frame. Experimental results show that the proposed approach outperforms several recent methods in coding artifact reduction.

AB - A compression artifacts removal algorithm that is adaptive to the artifact visibility level of the input video signal is proposed. The artifact visibility is determined per frame by the ratio of the accumulated gradient on the block edges to that of the remaining area. The filtering of each video frame is optimized using a least mean square mechanism which trains on pairs of target images and decompressed images of similar quality as the input frame. Experimental results show that the proposed approach outperforms several recent methods in coding artifact reduction.

M3 - Conference contribution

SP - 89

EP - 900

BT - Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference, March 31 2008-April 4 2008

T2 - 2008 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2008)

Y2 - 30 March 2008 through 4 April 2008

ER -

Quality adaptive trained filters for compression artifacts removal

Samenvatting

Congres

Toegang tot document

Vingerafdruk

Citeer dit