In this paper we focus on improving object detection performance in terms of recall. We propose a post-detection stage during which we explore the image with the objective of recovering missed detections. This exploration is performed by sampling object proposals in the image. We analyse four different strategies to perform this sampling, giving special attention to strategies that exploit spat...

We present an approach to function-based object recognition that reasons about the functional-ity of an object's intuitive parts. We extend the popular \recognition by parts" shape recognition framework to support \recognition by functional parts", by combining a set of functional primitives and their relations with a set of abstract volumetric shape primitives and their relations. Previous app...

Knowledge bases, data bases and object-oriented systems (referred to in the paper as Object-Centered systems) all rely on attributes as the main construct used to associate properties to objects; among these, a fundamental role is played by the so-called part-whole relation. The representation of such a structural information usually requires a particular semantics together with specialized inf...

This paper presents a method for object matching that uses local graphs called keygraphs instead of simple keypoints. A novel method to compare keygraphs was proposed in order to exploit their local structural information, producing better local matches. This speeds up an object matching pipeline, particularly using RANSAC, because each keygraph match contains enough information to produce a po...

In this paper, we present algorithms which allow an object-oriented querying of existing relational databases. Our goal is to provide an improved query interface for relational systems with better query facilities than SQL. This seems to be very important since, in real world applications, relational systems are most commonly used and their dominance will remain in the near future. To overcome ...

In this paper, we describe research into learning a model of the general 3D spatial relationships between objects found in human environments. Our approach trains a maximum entropy model using the qualitative spatial relationships between all the objects in several environments and produces a model of the underlying spatial regularities between the objects. It is tested as a purely spatial obje...

THE DESIGNATION "OBJECT RELATIONS THEORY" has been used with reference to a wide range of very different kinds of formulations: from Klein's rich and complex depiction of unconscious fantasy, to Fairbairn's highly abstract, schematic structural model, to Bowlby's ethologically-based theory of attachment, to Winnicott's epigramatic paradoxes and pithy observations about children, to Mahler's pow...

In this paper we investigate the object-oriented notion of subtyping in the context of behavioural formalisms. Subtyping in OO-formalisms is closely related to the concept of inheritance. The central issue in the choice of subtyping relations among classes is the principle of substitutability: an instance of the subtype should be usable wherever an instance of the supertype was expected. Depend...

This paper describes the construction of a qualitative spatial reasoning system based on the sensor data of a mobile robot. The spatial knowledge of the robot is formalised in three sets of axioms. First of all, axioms for relations between pairs of spatial regions are presented. Assuming the distance between regions as a primitive function in the language, the main purpose of this initial axio...

Representing parthood relations in ORM has received little attention, despite its added-value of the semantics at the conceptual level. We introduce a high-level taxonomy of types of meronymic and mereological relations, use it to construct a decision procedure to determine which type of part-whole role is applicable, and incrementally add mandatory and uniqueness constraints. This enables the ...