Delfina Sol Martinez Pandiani, Stefano De Giorgis, Fiorela Ciroku. This is the ontological module formalizing the schema of the Visual Genome model, aiming to represent the relations between objects in specific Regions of an Image. It implements some Ontology Design Patterns like Description and Situation and N-ary relation. It is furthermore aligned to DUL - Dolce Ultralight foundational Ontology and imports frame evocation classes from the Framester hub. 1.0 - Adding classes and properties. 1.1 - DUL import and alignement + frame evocation properties import from Framester schema. 1.2 - Introduced data properties aligned with DUL + refined classes axiomatization. 1.3 - URI control. 1.4 - Refined classes axiomatization. A catch-all object property, useful for alignment and querying purposes. It is declared as both transitive and symmetric, in order to reason an a maximal closure of associations between individuals associated with A relation between a Concept and an Entity classifies A generic, relative spatial location, holding between any entities. has location A schematic relation between any entities. has part A relation between entities and situations. has setting A relation between an information object and an Entity (including information objects). is about A relation between a Concept and an Entity. Equivalent to dul:isClassifiedBy. A generic, relative localization, holding between any entities. is location of A relation between any entities. is part of A relation between information objects and any Entity (including information objects). is reference of Any relation between concepts. is related to concept A relation between situations and entities. is setting for evokes evoked by evokes The relation between unary projections of a frame, and their ontological type, when given explicitly as a class in a knowledge representation language. It is a special case of the :subsumedUnder reification of subclass relation. onto type A subsumtion relation between two entities. subsumed under A subsumption relation between two meanings. subsumes Some ConceptualFrame is evoked by some Entity in the FRED graph produced by some Description of some DepictedRegion. evoked by associated with Equivalent to dul:classifies classifies Some ImageObject depicts some DepictedObject or DepictedRegion. depicts Some ConceptualFrame or WordNet 3.0 synset is evoked by an Entity. evoked by Some Entity evokes some ConceptualFrame or WordNet 3.0 synset. evokes Equivalent to dul:hasLocation. has location Equivalent to dul:hasPart. has part Some ObjectRelation takes some DepictedObject as object of the relationship. has relationship object Some ObjectRelation takes some DepictedObject as subject of the relationship. has relationship subject Equivalent to dul:hasSetting. has setting Some DepictedRegion includes (is the setting for) some DepictedObject. includes depicted object Equivalent to dul:includesObject. includes object Some DepictedRegion includes (is the setting for) some ObjectRelation. includes object relationship Equivalent to dul:isAbout. is about Equivalent to DUL:isClassifiedBy is classified by Some DepictedObject or DepictedRegion is depicted in some ImageObject. is depicted in Some DepictedObject is included (has setting) in some DepictedRegion. is depicted object included in Equivalent to dul:isLocationOf is location of Equivalent to dul:isObjectIncludedIn. is object included in Some ObjectRelation is included (has setting) in some DepictedRegion. is object relationship included in Equivalent to dul:isPartOf. is part of Equivalent to dul:isReferenceOf. is reference of Equivalent to dul:isRelatedToConcept. is related to concept Some DepictedObject is classified as the object of an ObjectRelation. is relationship object of Some DepictedObject is classified as the subject of an ObjectRelation. is relationship subject of Equivalent to dul:isSettingFor. is setting for Some DepictedObj, DepictedRel or ObjAttr is typed by a WordNet 3.0 synset. is typed by Some WordNet 3.0 synset types some DepictedObj, DepictedRel or ObjAttr. onto type Some WordNet 3.0 synset is subsumed under a ConceptualFrame. subsumed under Some ConceptualFrame subsumes some WordNet 3.0 synset. subsumes has data value has height Some Object has as attribute some string. has object attribute Some Conceptual Frame has as type some string such as "frame", "value" or "emotion". has frame type Some ImageObject has a URL path to a jpg provided by Visual Genome. has path URL A datatype property that encodes values for a Region, e.g. a float for the Region Height. has region data value has width has x coordinate has y coordinate Anything: real, possible, or imaginary, which some modeller wants to talk about for some purpose. Entity A piece of information, such as a musical composition, a text, a word, a picture, independently from how it is concretely realized. Information Object Any physical, social, or mental object, or a substance. Following DOLCE Full, objects are always participating in some event (at least their own life), and are spatially located. Object Any region in a dimensional space (a dimensional space is a maximal Region), which can be used as a value for a quality of an Entity . Region A view, consistent with ('satisfying') a Description, on a set of entities. It can also be seen as a 'relational context' created by an observer on the basis of a 'frame' (i.e. a Description). Situation Any Object that has a proper space region. The prototypical physical object has also an associated mass, but the nature of its mass can greatly vary based on the epistemological status of the object (scientifically measured, subjectively possible, imaginary). Social Object Any Region in a dimensional space that is used to localize an Entity. Space Region Frames as intended by Fillmore's frame semantics: the basic elements of semantic intepretation of natural language, independent from a specific lexicon (but not necessarily from a specific culture), necessarily evoked by any word, typically associated with a real world occurrence (situation) when evoked. When considered as multigrade predicates (n-ary relations, with role places and value positions within places), frame elements are binary projections of a multigrade predicate, where the first argument of the projection is always the (reified) event or situation occurring wrt to the evoked frame. Conceptual Frame The main class of the Framester schema. It is fully compliant to framenet:Frame, but extends it by providing alignments to the D&S and Semiotics ontology design patterns, and novel elements to deal with incomplete framal predicates, non-conceptual frames, etc. Frame Frames evoked by any (sense of a) word from a collection of words characterized by equivalent senses. Synset The Space Region physically occupied by some Entity in an ImageBox. i.e. the box that bounds some Entity parametrized by some coordinates, measured in pixels. Defined in the Visual Genome annotation methodology. Bounding Box To be completed Concept Conceptual Frame Any DepictedObject according to the Visual Genome annotation methodology. Depicted Object 2 A DepictedRegion in an Image. It is treated as a DUL:Situation and, in order to be meaningful, must include some ObjectRelation and a minimum of 2 DepictedObjects. Depicted Region Entity The total Space Region of some Image. Image Box An image object is a type of information object which has a location in a box of pixels (an ImageBox). It may depict DepictedObjects and DepictedRegions, and evoke ConceptualFrames. Image Object Information Object Object A relationship between two DepictedObjects, expressed by a predicate. Object Relationship Region Situation Social Object Space Region WN Synset Frame