Object (computer science)

About: Object (computer science) is a research topic. Over the lifetime, 106024 publications have been published within this topic receiving 1360115 citations. The topic is also known as: obj & Rq.

Method and apparatus for manipulating computer-based representations of objects of complex and unique geometry

Abstract: The present invention provides a method and apparatus for the manipulating computer-based representations of objects of complex and unique geometry. A computer acquires data describing an object and its surroundings, constructs a computer-based three dimensional model of the object from that data, superimposes an ideal geometry on the computer-based model, and alters the ideal geometry to fit the form and function required of the reproduction. The present invention presents two embodiments for manipulating computer-based representations of three dimensional objects.

MaskLab: Instance Segmentation by Refining Object Detection with Semantic and Direction Features

Abstract: In this work, we tackle the problem of instance segmentation, the task of simultaneously solving object detection and semantic segmentation. Towards this goal, we present a model, called MaskLab, which produces three outputs: box detection, semantic segmentation, and direction prediction. Building on top of the Faster-RCNN object detector, the predicted boxes provide accurate localization of object instances. Within each region of interest, MaskLab performs foreground/background segmentation by combining semantic and direction prediction. Semantic segmentation assists the model in distinguishing between objects of different semantic classes including background, while the direction prediction, estimating each pixel's direction towards its corresponding center, allows separating instances of the same semantic class. Moreover, we explore the effect of incorporating recent successful methods from both segmentation and detection (e.g., atrous convolution and hypercolumn). Our proposed model is evaluated on the COCO instance segmentation benchmark and shows comparable performance with other state-of-art models.

TOUCHDOWN: Natural Language Navigation and Spatial Reasoning in Visual Street Environments

Abstract: We study the problem of jointly reasoning about language and vision through a navigation and spatial reasoning task. We introduce the Touchdown task and dataset, where an agent must first follow navigation instructions in a Street View environment to a goal position, and then guess a location in its observed environment described in natural language to find a hidden object. The data contains 9326 examples of English instructions and spatial descriptions paired with demonstrations. We perform qualitative linguistic analysis, and show that the data displays a rich use of spatial reasoning. Empirical analysis shows the data presents an open challenge to existing methods.

Method and apparatus for dynamic update of an existing object in an object editor

Abstract: User interface objects are stored in a user interface object archive which is a database physically located in the shared library of an associated application program. In order to facilitate "localization", or preparation of an application developed in one language for use in an "area" or a locale which uses another language, the user interface objects are stored in a hierarchical locale tree within the archive. All objects are stored in the base or root of the hierarchy, but only those objects which require a translation are stored in an area associated with a more specific locale. At runtime, a complete collection of objects is assembled by starting at the desired locale and proceeding up the hierarchy level-by-level. Translated objects at lower levels of the hierarchy "override" those at higher levels so that the most complete translations of each object are obtained during this search. The user interface objects which are stored in an archive are actually created via a predefined "constructor" program, and, in order to allow newly-created user interface objects to use the predesigned constructor program, each user interface object is contained in a special "escort" object that interfaces with the constructor program. A user can edit an existing object, and change its class. This results in a new object of the new class being created. The existing attributes of the old object are transferred to the new object. Only the attributes known to the common base class or classes of the two objects are transferred. The new object is substituted for the old object, and the old object is deleted.

Dynamic object- driven database manipulation and mapping system

Abstract: The present invention provides a system and method for dynamic object-driven database manipulation and mapping system which relates in general to correlating or translating one type of database to another type of database or to an object programming application. Correlating or translating involves relational to object translation, object to object translation, relational to relational, or a combination of the above. Thus, the present invention is directed to dynamic mapping of databases to selected objects. Also provided are systems and methods that optionally include caching components, security features, data migration facilities, and components for reading, writing, interpreting and manipulating XML and XMI data files.