Real-world, multiple-typed objects are often interconnected, forming heterogeneous information networks. A major challenge for link-based clustering in such networks is its potent...
— In this paper, we consider the problem of how background knowledge about usual object arrangements can be utilized by a mobile robot to more efficiently find an object in an ...
We present an object recognition system that locates an object, identifies its parts, and segments out its contours. A key distinction of our approach is that we use long, salien...
In this paper, we present our approach and architecture for fault diagnosis and self-healing of interpreted objectoriented applications. By combining aspect-oriented programming, ...
A. Reza Haydarlou, Benno J. Overeinder, Frances M....
Visual attributes expose human-defined semantics to object recognition models, but existing work largely restricts their influence to mid-level cues during classifier training....