Abstract. Recognizing group activities is challenging due to the difficulties in isolating individual entities, finding the respective roles played by the individuals and representing the complex interactions among the participants. Individual actions and group activities in videos can be represented in a common framework as they share the following common feature: both are composed of a set of low-level features describing motions, e.g., optical flow for each pixel or a trajectory for each feature point, according to a set of composition constraints in both temporal and spatial dimensions. In this paper, we present a unified model to assess the similarity between two given individual or group activities. Our approach avoids explicit extraction of individual actors, identifying and representing the inter-person interactions. With the proposed approach, retrieval from a video database can be performed through Query-by-Example; and activities can be recognized by querying videos containing known activities. The suggested video matching process can be performed in an unsupervised manner. We demonstrate the performance of our approach by recognizing a set of human actions and football plays. 1

behavior classification

Abstract—Deployments of camera security systems are capable of capturing long data sequences about human activity. This paper deals with processing of detected sequences at a more macroscopic level to detect chains of events based on a prior given specification. In our problem, sensed interactions between people are modeled as sequences of pairwise events that are interleaved with other interactions taking place in the background. We formulate the problem as an isomorphic subgraph matching problem and solve it to detect a chain of events, its participants and their roles. We further evaluate our solution in the presence of background interference from other interactions and give analytical and empirical results about the performance of our algorithm. I.

Resumen: El objetivo de nuestra investigación es el de generar resúmenes de películas a partir de textos colaterales, capturando el contenido semántico, estructura narrativa y líneas clave de los diálogos de la película. Nuestra hipótesis es que se pueden generar de forma eficiente resúmenes en texto de películas mediante el empleo de técnicas de resumen automático sobre textos colaterales: subtítulos, descripciones del audio y guiones de postproducción. En caso de disponer de códigos de tiempo, entonces podemos generar también resúmenes en vídeo a partir de dichos resúmenes en texto. En este estudio inicial construimos los resúmenes seleccionando las diez tomas de la película original que contienen la mayor proporción de palabras clave. Definimos las palabras clave de dos formas: como palabras de frecuencia media, ya que son las palabras de frecuencia media en un texto las que contienen la mayor parte de la información acerca del contenido de dicho texto; y como entidades nombradas derivadas del reparto de la película. Ye et al. (2007) sostienen que la calidad de un resumen puede evaluarse en base a cuántos de los conceptos principales del texto original se conservar en el resumen. Hemos comprobado que esta aproximación a la evaluación de resúmenes obtiene resultados más favorables que la

Today’s organisations require techniques for automated transformation of their large data volumes into operational knowledge. This requirement may be addressed by employing event recognition systems that detect events/activities of special significance within an organisation, given streams of ‘low-level ’ information that is very difficult to be utilised by humans. Consider, for example, the recognition of attacks on nodes of a computer network given the TCP/IP messages, the recognition of suspicious trader behaviour given the transactions in a financial market, and the recognition of whale songs given a symbolic representation of whale sounds. Various event recognition systems have been proposed in the literature. Recognition systems with a logic-based representation of event structures, in particular, have been attracting considerable attention, because, among others, they exhibit a formal, declarative semantics, they have proven to be efficient and scalable, and they are supported by machine learning tools automating the construction and refinement of event structures. In this paper we review representative approaches of logic-based event recognition and discuss open research issues of this field. We illustrate the reviewed approaches with the use of a real-world case study: event recognition for city transport management. 1