ICME 2005
117 papers
The 2005 edition of the IEEE International Conference on Multimedia and Expo (ICME) featured 139 papers covering a range of multimedia research topics. A primary area of focus involved video coding and compression standards, specifically the H.264/AVC standard, its scalable extensions, and joint rate control for multiple sequences. Research in computer vision and pattern recognition was also prominent, with studies addressing face detection using local descriptors, audio-visual speech recognition through multi-stream Hidden Markov Models, and layer segmentation for automated visual surveillance.
Prolific authors contributing to the proceedings included Hong-Jiang Zhang, Xian-Sheng Hua, Hari Sundaram, and Nicu Sebe. Additional research themes encompassed image authentication via feature-amplified watermarking and the detection of salient visual attention using adaptive local context suppression. The year’s contributions also explored multimedia applications such as video booklets. Collectively, these works represent a focus on enhancing video transmission efficiency and improving automated analysis techniques for audio-visual data.
Top Research Directions
Research directions most represented by ICME 2005 papers.
| The Clean Signal Penalty in Robust Speech Models | 1 |
| Distributed Architectures for Speech and Audio Processing | 1 |
| Overcoming the Stationary Noise Assumption in Signal Enhancement | 1 |
Top Authors
Impactful Papers
Papers from ICME 2005 that shaped research clusters.
Foundational
Events Detection for an Audio-Based Surveillance System
foundational
The Clean Signal Penalty in Robust Speec
Influential
Events Detection for an Audio-Based Surveillance System
influential
Distributed Architectures for Speech and
Events Detection for an Audio-Based Surveillance System
influential
Overcoming the Stationary Noise Assumpti
Papers
« prev Page 3/3