ACM Multimedia 2013
145 papers
The 2013 edition of ACM Multimedia featured 216 papers, with a prominent focus on affective computing and high-dimensional data analysis. Research directions included bridging the affective gap in computer vision and exploring continuous dimensional models to analyze the asymmetry of valence and arousal. Methodological trends centered on supervised dimensionality reduction for feature extraction and modeling structured relationships within label spaces to improve inference.
Notable work during this year addressed diverse applications such as multimedia event detection through fine-grained labeling and visual business recognition using multimodal approaches. Other contributions explored location recognition via geo-informative attributes and non-linear media access in peer-to-peer networks. Research also examined image representation in sub-semantic spaces, moving beyond traditional bag-of-words models. Prolific contributors included researchers such as Qi Tian, Shuicheng Yan, and Shih-Fu Chang, whose work spanned areas from superpixel segmentation for scene recognition to non-reference audio quality assessment for live recordings.
Top Research Directions
Research directions most represented by ACM Multimedia 2013 papers.
Top Authors
Impactful Papers
Papers from ACM Multimedia 2013 that shaped research clusters.
Foundational
Learning articulated body models for people re-identification
foundational
3D Body Geometry for Invariant Person Re
Structured exploration of who, what, when, and where in heterogeneous multimedia news sources
foundational
Competitive Matching for Audio-Visual Ac
Influential
Correlated-spaces regression for learning continuous emotion dimensions
influential
Modeling the Temporal Evolution of Facia
Object co-segmentation via discriminative low rank matrix recovery
influential
Robust Separation via Low Rank and Spars
Multi-factor segmentation for topic visualization and recommendation: the MUST-VIS system
influential
Scaling Video Moment Retrieval across Ma
Learning articulated body models for people re-identification
influential
The Upstream Pose Estimation Bottleneck
Non-rigid target tracking based on 'flow-cut' in pair-wise frames with online hough forests
influential
Robust Visual Tracking Under Appearance
Papers
« prev Page 3/3