Abstract: The reproduction and modeling of natural phenomena using computer graphics is used in a wide range of fields. This requires a great deal of work on the part of the producer. Procedural techniques are an effective means of supporting this process, and this paper focuses on the generation and modeling of branching structures. First, this paper extends an existing branching structure generation algorithm, space colonization, by changing the positions of the points that make up the segments of the Out-Tree generated in 3D space. By using an induced vector field rotated by quaternions around the normal vector of the starting point of the Out-Tree generation, the algorithm can change the coordinate values of newly generated points, thereby allowing the shape of the branching structure to be manipulated by numerical parameters. The next step is to apply the Controllable Field Space Colonization as a framework for modeling. The application domain of the algorithm is determined by the simple model input by the user. The input model is converted to sparse volume data, and the attraction points are placed on the surface or inside the input model. This method can not only generate tree models similar to the L-System and existing space colonization, but also represent frost, lightning, blood vessels, rivers and mountains. As the future work, includes optimization to enable automatic input of numerical values for each generated model, and correction of polygon flipping due to normal vector errors.Abstract: The reproduction and modeling of natural phenomena using computer graphics is used in a wide range of fields. This requires a great deal of work on the part of the producer. Procedural techniques are an effective means of supporting this process, and this paper focuses on the generation and modeling of branching structures. First, this paper extend...Show More
Abstract: The algorithm of effective geometric hashing of the facial feature hyperspace for the accelerated search of the most similar face descriptors by their cosine similarity is described in the present study. The algorithm includes 6 required stages of processing descriptors extracted by a neural network from face images. The first stage is filtration of the descriptor database by selecting the most representative descriptor for each person from the set of descriptors corresponding to his/her different face images. The second stage is evaluation of a number of statistical values for all the components of the selected descriptors. The third stage is intermediate hashing through quantization of every descriptor component value so that almost the same quantity of descriptors corresponds to any quantum number. The fourth stage is statistical processing of the descriptor database to determine the most discriminative descriptor key components and their hierarchy. The fifth stage is calculation of the descriptor hash code for every most representative descriptor from the considered database. The sixth final stage is a special cataloging of data in the form of a multi-tiered directory ordered by the hash codes. The search acceleration is achieved through sparse processing of the whole directory when the hash code obtained for the requested person descriptor acts as a very selective search filter. The developed algorithm always provides the same absolute accuracy as the brute-force search. Through the example of the LFW dataset consideration, the average search acceleration by about 100 times is achieved under conditions that the descriptors have been extracted by a neural network trained on the WiderFace dataset with application of the additive angular margin loss function.Abstract: The algorithm of effective geometric hashing of the facial feature hyperspace for the accelerated search of the most similar face descriptors by their cosine similarity is described in the present study. The algorithm includes 6 required stages of processing descriptors extracted by a neural network from face images. The first stage is filtration o...Show More