DO YOU WANT TO KNOW MORE ABOUT 3D TURTLES? ENTER HERE
3D TURTLES
3D TURTLES
3D TURTLES
3D TURTLES
3D TURTLES
3DFORM-ID: AN IDENTIFICATION DEVICE CONTAINING THE TRIDIMENSIONAL SHAPE DESCRIPTION OF A REAL WORLD´S OBJECT AND A PROCEEDING RELATED TO THE USE OF THIS IDENTIFICATION DEVICE
For a computer based system, recognizing real world objects can be computationally expensive, or impossible in practical terms, using the current state-of-the-art input methods and recognition procedures. The device and proceeding described in this paper solves this problem, as any real world´s object can be identified quickly and exactly by a computer based system, if the real world´s object to be identified has been previously provided with this identification device containing the tridimensional form description of the real world´s object, attached or implemented on it, and if a computer based system have also available the accurate tools to detect, to read and to decode this identification device invention. This device will contains the tridimensional shape description and also the physical properties of the object where the referred device is attached or implemented, then allowing to any capable computer based system to build tridimensional scenes quickly and predict how these objects can interact between them in the created virtual scene and then also predict how their real world´s counterparts would act or how could be them manipulated in the real world. In this document, this device and the related proceedings will be referred as "3DFORM-ID DEVICE" and "3DFORM-ID PROCEEDING". Some previous approaches [1] to describe the geometry of a real world object has been done using multiple tags spreaded over the surface of an object. But in this current paper, the object´s geometrical description is stored in a single tag to be read by an external system, simplifying enormously the 3D recognition process of real world´s objects. The novelty and impact of the 3DFORM-ID device has been recently remarked by some worlwide RFID related publications [2], [3]. The 3DFORM-ID device has been also filled as a patent by this paper´s author [4].
3DLABEL-ID : AN IDENTIFICATIVE LABEL THAT CONTAINS GRAPHIC AND/OR ELECTROMAGNETICAL REFERENCES FOR RELATING A TRIDIMENSIONAL SHAPE TO A REAL WORLD OBJECT
The use of code bars and radiofrecuency devices (RFIDs) allows the identification of a product using a number that links the identified object with data located externally. Using a proceeding based in those described at the european patent filing EP06381005.5, explaining a device named 3DFORM-ID that contains a referenced tridimensional shape of a real world object, the proposed identificative label allows relating the identified object with an internal or external file that contains the shape description of the identified object and the necessary spatial references to obtain their spatial position and orientation accordingly to a detection system, using graphical or electromagnetical references included in one or more identificative labels printed or adhered over the identified object.
TUE: TRIDIMENSIONAL UNDERSTANDING ENGINES
The method described in this paper consists on recreating an emulator to allow true understanding for a computer, by using tools that already exist on the field of tridimensional computer graphics. The method is based on using shapes representing tridimensional objects from the reality and actions that affect these objects, all of it in an euclidean reference system that is active in the memory of the computer. The basic idea is that computers do not understand human natural language because language is grounded on the basis of a symbolic system that makes no sense for a computer, as is said in the famous “chinesse room” paradox [1], unless the computer is previously provided with a code that allows to the computer putting this symbolic system in connection with a reference system where the machine may have its own representation of the terms provided with respect to itself. Using 3D techniques aren´t a new approach to Artificial Intelligence. Since the dawn of computer graphics, thousands of propossals have been done around the use of 3D procedures in this area. But on despite of these efforts, few true advances to AI have been obtained. This has been on this way mainly as a result of three factors: 1.The difficulty to recognize real world objects. 2. The use of 3D models as symbols. 3. mixing abstract and real, physical concepts. Language is a tool that allows us to articulate our thoughts, whether for ourselves or to communicate with others. Nevertheless, words are not thoughts in themselves, but rather a depiction of said thought by means of a code formed by symbols, whether verbal, written or gesticulated. They are a final “rendering” of something deeper that we call thought. Some authors [2], mainly involved in Language of Thought Hipotesis (LOTH), have named “mindish” or “mentalish” to this internal, supraverbal language that precedes expression through words. Starting on the assumption that our brain works, at a deep level, with a language that is based not on words, but rather in rough aproximations elaborated with morphologies and actions previously stored in some way in our brain memory, we build an empirical method that is suitable for the specific features of a computer with graphic tridimensional capacity. The method in question is called TUE, abbreviation for Tridimensional Understanding Engine. More specifically we call T3UE to a system focused to understanding words with no time variations, like “stone” or “house”, or understanding a syntactic description of a complex but steady scene. We call T4UE (Tetradimensional Understanding Engine) to a system aimed at understanding complex sentences that involves objects and actions which take place in time.