Fast software processing of motion JPEG video
MULTIMEDIA '94 Proceedings of the second ACM international conference on Multimedia
Inner-block operations on compressed images
Proceedings of the third ACM international conference on Multimedia
Image and Video Compression Standards: Algorithms and Architectures
Image and Video Compression Standards: Algorithms and Architectures
A fast approximate algorithm for scaling down digital images in the DCT domain
ICIP '95 Proceedings of the 1995 International Conference on Image Processing (Vol.2)-Volume 2 - Volume 2
On the extraction of DC sequence from MPEG compressed video
ICIP '95 Proceedings of the 1995 International Conference on Image Processing (Vol.2)-Volume 2 - Volume 2
Rate conversion of MPEG coded video by re-quantization process
ICIP '95 Proceedings of the 1995 International Conference on Image Processing (Vol. 3)-Volume 3 - Volume 3
A fast algorithm for DCT-domain inverse motion compensation
ICASSP '96 Proceedings of the Acoustics, Speech, and Signal Processing, 1996. on Conference Proceedings., 1996 IEEE International Conference - Volume 04
A transform domain approach to spatial domain image scaling
ICASSP '96 Proceedings of the Acoustics, Speech, and Signal Processing, 1996. on Conference Proceedings., 1996 IEEE International Conference - Volume 04
Manipulation and compositing of MC-DCT compressed video
IEEE Journal on Selected Areas in Communications
Hi-index | 0.00 |
The emergence of compression standards such as JPEG, MPEG, H.320 and H.324 has enabled many consumer and business multimedia applications wherein the multimedia content is disseminated in their compressed form. Many applications however require processing of the multimedia content prior to presentation. Traditional approaches for processing rely on first decompressing the bitstreams and then applying the desired processing function. In this paper we describe recent developments that have led to processing techniques that can be directly implemented on an intermediate representation of the compressed domain information. By avoiding the computationally expensive decompression and recompression processes needed in traditional spatial/temporal domain based techniques, mediaprocessing in the compressed domain achieves 2-4 fold speedup for typical processing functions such as image downscaling, bitrate conversions, interframe to intra- frame conversions.