• Journal of KIISE
• /
• v.43 no.7
• /
• pp.795-800
• /
• 2016

In image classification, recent CNNs compete with human performance. However, there are limitations in more general recognition. Herein we deal with indoor images that contain too much information to be directly processed and require information reduction before recognition. To reduce the amount of data processing, typically variational inference or variational Bayesian methods are suggested for object detection. However, these methods suffer from the difficulty of marginalizing over the given space. In this study, we propose an image-text integrated recognition system using active vision based on Spatial Transformer Networks. The system attempts to efficiently sample a partial region of a given image for a given language information. Our experimental results demonstrate a significant improvement over traditional approaches. We also discuss the results of qualitative analysis of sampled images, model characteristics, and its limitations.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.6
• /
• pp.227-233
• /
• 2019

Recently, along with the recent development of deep learning technique, neural networks are achieving success in computer vision filed. Convolutional neural network have shown outstanding performance in not only for a simple image classification task, but also for tasks with high difficulty such as object segmentation and detection. However many such deep learning models are based on supervised-learning, which requires more annotation labels than image-level label. Especially image semantic segmentation model requires pixel-level annotations for training, which is very. To solve these problems, this paper proposes a weakly-supervised semantic segmentation method which requires only image level label to train network. Existing weakly-supervised learning methods have limitations in detecting only specific area of object. In this paper, on the other hand, we use multi-classifier deep learning architecture so that our model recognizes more different parts of objects. The proposed method is evaluated using VOC 2012 validation dataset.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.2
• /
• pp.349-356
• /
• 2020

This paper introduces the implementation of a driver's condition warning system using eye aspect ratio to prevent a car accident. The proposed driver's condition warning system using eye aspect ratio consists of a camera, that is required to detect eyes, the Raspberrypie that processes information on eyes from the camera, buzzer and vibrator, that are required to warn the driver. In order to detect and recognize driver's eyes, the histogram of oriented gradients and face landmark estimation based on deep-learning are used. Initially the system calculates the eye aspect ratio of the driver from 6 coordinates around the eye and then gets each eye aspect ratio values when the eyes are opened and closed. These two different eye aspect ratio values are used to calculate the threshold value that is necessary to determine the eye state. Because the threshold value is adaptively determined according to the driver's eye aspect ratio, the system can use the optimal threshold value to determine the driver's condition. In addition, the system synthesizes an input image from the gray-scaled and LAB model images to operate in low lighting conditions.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.5
• /
• pp.537-542
• /
• 2021

This paper improves the possibility of recognizing fast-moving objects through the YOLO (You Only Look Once) deep learning recognition model in an application environment for object recognition in images. The purpose was to study the method of collecting semantic data through processing. In the recognition model, the moving object recognition error was identified as unrecognized because of the difference between the frame rate of the camera and the moving speed of the object and a misrecognition due to the existence of a similar object in an environment adjacent to the object. To minimize the recognition errors by compensating for errors, such as unrecognized and misrecognized objects through the proposed data collection method, and applying vision processing technology for the causes of errors that may occur in images acquired for sports (tennis games) that can represent real similar environments. The effectiveness of effective secondary data collection was improved by research on methods and processing structures. Therefore, by applying the data collection method proposed in this study, ordinary people can collect and manage data to improve their health and athletic performance in the sports and health industry through the simple shooting of a smart-phone camera.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.6
• /
• pp.1321-1330
• /
• 2023

This study attempts to address the problem of 3D pose estimation for multiple human objects through a single image generated during the character development process that can be used in augmented reality. In the existing top-down method, all objects in the image are first detected, and then each is reconstructed independently. The problem is that inconsistent results may occur due to overlap or depth order mismatch between the reconstructed objects. The goal of this study is to solve these problems and develop a single network that provides consistent 3D reconstruction of all humans in a scene. Integrating a human body model based on the SMPL parametric system into a top-down framework became an important choice. Through this, two types of collision loss based on distance field and loss that considers depth order were introduced. The first loss prevents overlap between reconstructed people, and the second loss adjusts the depth ordering of people to render occlusion inference and annotated instance segmentation consistently. This method allows depth information to be provided to the network without explicit 3D annotation of the image. Experimental results show that this study's methodology performs better than existing methods on standard 3D pose benchmarks, and the proposed losses enable more consistent reconstruction from natural images.

• Korean Journal of Construction Engineering and Management
• /
• v.23 no.3
• /
• pp.45-55
• /
• 2022

BIM models allow building spaces to be instantiated and recognized as unique objects independently of model elements. These instantiated spaces provide the required semantics that can be leveraged for building code checking, energy analysis, and evacuation route analysis. However, theses spaces or rooms need to be designated manually, which in practice, lead to errors and omissions. Thus, most BIM models today does not guarantee the semantic integrity of space designations, limiting their potential applicability. Recent studies have explored ways to automate space allocation in BIM models using artificial intelligence algorithms, but they are limited in their scope and relatively low classification accuracy. This study explored the use of Graph Convolutional Networks, an algorithm exclusively tailored for graph data structures. The goal was to utilize not only geometry information but also the semantic relational data between spaces and elements in the BIM model. Results of the study confirmed that the accuracy was improved by about 8% compared to algorithms that only used geometric distinctions of the individual spaces.