TY - GEN

T1 - The New MPEG-4/FAMC Standard for Animated 3D Mesh Compression

AU - Mamou, K.

AU - Stefanoski, N.

AU - Kirchhoffer, H.

AU - Zaharia, T.

AU - Preteux, F.

AU - Marpe, D.

AU - Ostermann, J.

AU - Müller, K.

PY - 2008

Y1 - 2008

N2 - This paper presents a new compression technique for 3D dynamic meshes, referred to as FAMC - Frame-based Animated Mesh Compression, recently promoted within the MPEG-4 standard as Amendement 2 of part 16 (AFX - Animation Framework eXtension). The FAMC approach combines a model-based motion-compensation strategy with transform/predictive coding of residual errors. First, a skinning motion-compensation model is automatically derived from a frame-based representation. Subsequently, either 1) DCT/lifting wavelets or 2) layer-based predictive coding is employed to exploit remaining spatio-temporal correlations in the residual signal. Both motion model parameters and residual signal components are finally encoded by using context-based adaptive binary arithmetic coding (CABAC). The proposed FAMC encoder offers high compression performance with gains of 60% in terms of bit-rate savings relative to previous MPEG-4 technology and of 20% to 40% relative to state-of-the-art techniques. FAMC is well suited for compressing both geometric and photometric (normal vectors, colors...) attributes. In addition, FAMC also supports a rich set of functionalities including streaming, scalability (spatial, temporal and quality) and progressive transmission.

AB - This paper presents a new compression technique for 3D dynamic meshes, referred to as FAMC - Frame-based Animated Mesh Compression, recently promoted within the MPEG-4 standard as Amendement 2 of part 16 (AFX - Animation Framework eXtension). The FAMC approach combines a model-based motion-compensation strategy with transform/predictive coding of residual errors. First, a skinning motion-compensation model is automatically derived from a frame-based representation. Subsequently, either 1) DCT/lifting wavelets or 2) layer-based predictive coding is employed to exploit remaining spatio-temporal correlations in the residual signal. Both motion model parameters and residual signal components are finally encoded by using context-based adaptive binary arithmetic coding (CABAC). The proposed FAMC encoder offers high compression performance with gains of 60% in terms of bit-rate savings relative to previous MPEG-4 technology and of 20% to 40% relative to state-of-the-art techniques. FAMC is well suited for compressing both geometric and photometric (normal vectors, colors...) attributes. In addition, FAMC also supports a rich set of functionalities including streaming, scalability (spatial, temporal and quality) and progressive transmission.

KW - AFX

KW - Animation compression

KW - CABAC

KW - Dynamic mesh compression

KW - Mesh compression

KW - MPEG-4

UR - http://www.scopus.com/inward/record.url?scp=50949121113&partnerID=8YFLogxK

U2 - 10.1109/3DTV.2008.4547817

DO - 10.1109/3DTV.2008.4547817

M3 - Conference contribution

AN - SCOPUS:50949121113

SN - 9781424417551

T3 - 3DTV Conference

SP - 97

EP - 100

BT - 2008 3DTV-Conference

T2 - 2008 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video

Y2 - 28 May 2008 through 30 May 2008

ER -