Katerina Pastra

Learn More
We present a text-based approach for the automatic indexing and retrieval of digital photographs taken at crime scenes. Our research prototype, SOCIS, goes beyond keyword-based approaches and methods that extract syntactic relations from captions; it relies on advanced Natural Language Processing techniques in order to extract relational facts. These(More)
In this paper, we investigate whether reusing existing grammars for NE recognition instead of creating them from scratch is a viable solution to time constraints in developing grammars. We discuss three possible factors that hinder grammar reuse and we present our corresponding empirical results, that encourage more widespread use of valuable existing(More)
In this paper we attempt to apply the IBM algorithm, BLEU, to the output of four different summarizers in order to perform an intrinsic evaluation of their output. The objective of this experiment is to explore whether a metric, originally developed for the evaluation of machine translation output, could be used for assessing another type of output(More)
This paper presents work on text-based photograph indexing and retrieval for crime investigation, an application domain where efficient querying of large crime-scene photograph databases is of crucial importance. Automating this task will change current police practices considerably, by bringing ‘intelligence’ to crime support information systems. The(More)
A good data corpus lies at the heart of progress in both perceptual/cognitive science and in computer vision. While there are a few datasets that deal with simple actions, creating a realistic corpus for complex, long action sequences that contains also human-human interactions has so far not been attempted to our knowledge. Here, we introduce such a corpus(More)
The growing demand for intelligent multimedia systems has led to the development of various multimodal resources and corresponding annotation schemes and processing tools. In this paper, we argue that there is a striking lack of multimodal corpora capturing the association and interaction of visual and linguistic data. We relate this research lacuna to(More)
first find a crime scene, the objects and subjects as well as their spatial relations and conditions, all are crucial for collecting evidence and for drawing conclusions during crime investigation. Although investigating a crime is a time-consuming process, the crime scene cannot be preserved for long: life must take again its normal course, objects must be(More)
Natural language use, acquisition, and understanding takes place usually in multisensory and multimedia communication environments. Therefore, for one to model language in its interaction and integration with sensorimotor experiences, one needs a representative corpus of such interplay. In this paper, we will present the first corpus of language use and(More)
Though everyday interaction is predominantly multimodal, a purpose-developed framework for describing the semantic interplay between verbal and non-verbal communication is still lacking. This lack not only indicates one’s poor understanding of multimodal human behaviour, but also weakens any attempt to model such behaviour computationally. In this article,(More)