The tight coupling between language design and language processing in traditional language processing tools is avoided by model-based parser generators such as ModelCC. By decoupling language specification from language processing, ModelCC avoids the limitations imposed by traditional parser generators, which constrain language designers to specific kinds of grammars. Apart from providing an alternative approach to language specification, ModelCC incorporates reference resolution within the parsing process. Instead of returning mere abstract syntax trees, ModelCC is able to obtain abstract syntax graphs from its input string. Moreover, such abstract syntax graphs are not restricted to directed acyclic graphs, since ModelCC supports anaphoric, cataphoric, and recursive references.

Model-based parser generators such as ModelCC effectively decouple language design from language processing. ModelCC allows the specification of the abstract syntax model of a language as a set of language elements and their relationships. ModelCC provides the necessary mechanisms to specify the mapping from the abstract syntax model (ASM) to a concrete syntax model (CSM). This mapping can be specified as a set of metadata annotations on top of the abstract syntax model itself or by means of a domain-specific language (DSL). Using a domain-specific language to specify the mapping from abstract to concrete syntax models allows the definition of different concrete syntax models for the same abstract syntax model. In this paper, we describe the ModelCC domain-specific language for ASM-CSM mappings and we showcase its capabilities by using the ModelCC ASM-CSM DSL to define itself.

Syntax-directed translation tools require the specification of a language by means of a formal grammar. This grammar must also conform to the specific requirements of the parser generator to be used. Software engineers then annotate the resulting grammar with semantic actions for the resulting system to perform its desired functionality. Whenever the input text format is modified, the grammar has to be updated and the subsequent changes propagate throughout the entire language processing tool chain. Moreover, if several applications use the same language, multiple copies of the same language specification have to be maintained in sync, since language specification (i.e. the grammar) is tightly coupled to language processing (i.e. the semantic actions that annotate that grammar). In this paper, we introduce ModelCC, a model-based parser generator that decouples language specification from language processing, hence avoiding the aforementioned problems that are caused by grammar-driven parser generators. ModelCC receives a conceptual model as input, along with constraints that annotate it. It is then able to create a parser for a given textual representation format, so that the generated parser fully automates the instantiation of the desired conceptual model. ModelCC is also a powerful tool for the design of domain-specific languages.

ModelCC is a model-based parser generator. Model-based parser generators decouple language specification from language processing. This model-driven approach avoids the limitations imposed by parser generators whose language specifications must conform to specific grammar constraints. Moreover, ModelCC supports reference resolution within the language specification. Therefore, it does not parse just trees but it can also efficiently deal with abstract syntax graphs. These graphs can even include cycles (i.e. they are not constrained to directed acyclic graphs).

Motion tracking is a critical task in many computer vision applications. Existing motion tracking techniques require either a great amount of knowledge on the target object or specific hardware. These requirements discourage the wide spread of commercial applications based on motion tracking. In this paper, we present a novel three degrees of freedom motion tracking system that needs no knowledge on the target object and that only requires a single low-budget camera that can be found installed in most computers and smartphones. Our system estimates, in real time, the three-dimensional position of a nonmodeled unmarked object that may be nonrigid, nonconvex, partially occluded, self-occluded, or motion blurred, given that it is opaque, evenly colored, enough contrasting with the background in each frame, and that it does not rotate. Our system is also able to determine the most relevant object to track in the screen. Our proposal does not impose additional constraints, therefore it allows a market-wide implementation of applications that require the estimation of the three position degrees of freedom of an object.