MCO Object Ontology

Author(s):  
Dean S. Hartley III
Keyword(s):  
Author(s):  
Vasileios Mezaris ◽  
Ioannis Kompatsiaris ◽  
Michael G. Strintzis

2019 ◽  
Vol 27 (12) ◽  
pp. 144-157 ◽  
Author(s):  
M. B. Sapunov ◽  
A. A. Polonnikov

The paper focuses on the analysis of ontological, epistemological and pedagogical terms of changes in education, first of all – on the status of an academic subject. From the ontological perspective, they are related to transfer from “object” ontology to communication-and-activity one, in terms of epistemological approach – from naturalism to transcendentalism, whereas with regard to pedagogical perspective – from autocratic-disciplinary organization of the academic process to a student-oriented pattern. The first part of the article describes academic subject’s functions in the process of educational reproduction. An academic subject is interpreted not so much from the perspective of the knowledge it contains, but as a complex linguistic code which organizes and regulates educational interaction. Its basics and structure, a mechanism of constituting educational reality are described, as well as the design specifics that hamper changes in education. The authors dwell on the distinction between an academic and scholarly subject. Part two of the article contains criticism of an academic subject practice in the university education. The central event here is attributed to differentiation and diversification of the form of academic subject, disintegration of its integrity into local autonomous linguistic fields. The conclusion formulates the idea how to overcome an academic subject crisis, which heart is discursive transformation of its representation practice. Based on Gilles Deleuze’s ideas, the authors consider the transformation of discursive practices in which an academic subject is embodied to be the condition for education change.


Author(s):  
Eric Wang ◽  
Yong Se Kim

We present an ontology of objects, relations among objects, and generic shape representation that supports form-function reasoning. By reasoning from the generic functions of objects to their geometric shape requirements, we deduce the generic shape representation of everyday objects. This is a complex kind of reasoning that combines diverse knowledge sources and principles. We model the results of this reasoning process as a justification graph of individual reasoning steps, which explicitly links the attributes of objects and their relations to the corresponding geometric shape elements. This object ontology uses OWL Full metamodeling techniques to achieve the necessary level of expressiveness while maintaining a generic representation. We give an example for the Table class, showing its decomposition into functions, features, and relations, and its form-function reasoning process.


2019 ◽  
Vol 2019 ◽  
pp. 1-40
Author(s):  
Ngoc Q. Ly ◽  
Tuong K. Do ◽  
Binh X. Nguyen

Object retrieval plays an increasingly important role in video surveillance, digital marketing, e-commerce, etc. It is facing challenges such as large-scale datasets, imbalanced data, viewpoint, cluster background, and fine-grained details (attributes). This paper has proposed a model to integrate object ontology, a local multitask deep neural network (local MDNN), and an imbalanced data solver to take advantages and overcome the shortcomings of deep learning network models to improve the performance of the large-scale object retrieval system from the coarse-grained level (categories) to the fine-grained level (attributes). Our proposed coarse-to-fine object retrieval (CFOR) system can be robust and resistant to the challenges listed above. To the best of our knowledge, the new main point of our CFOR system is the power of mutual support of object ontology, a local MDNN, and an imbalanced data solver in a unified system. Object ontology supports the exploitation of the inner-group correlations to improve the system performance in category classification, attribute classification, and conducting training flow and retrieval flow to save computational costs in the training stage and retrieval stage on large-scale datasets, respectively. A local MDNN supports linking object ontology to the raw data, and an imbalanced data solver based on Matthews’ correlation coefficient (MCC) addresses that the imbalance of data has contributed effectively to increasing the quality of object ontology realization without adjusting network architecture and data augmentation. In order to evaluate the performance of the CFOR system, we experimented on the DeepFashion dataset. This paper has shown that our local MDNN framework based on the pretrained NASNet architecture has achieved better performance (14.2% higher in recall rate) compared to single-task learning (STL) in the attribute learning task; it has also shown that our model with an imbalanced data solver has achieved better performance (5.14% higher in recall rate for fewer data attributes) compared to models that do not take this into account. Moreover, MAP@30 hovers 0.815 in retrieval on an average of 35 imbalanced fashion attributes.


Sign in / Sign up

Export Citation Format

Share Document