• 한국멀티미디어학회논문지
• /
• 제9권6호
• /
• pp.693-699
• /
• 2006

Object-oriented databases (OODBs) have been adopted for managing non-standard applications such as computer-aided design (CAD), office document management and many multimedia applications. One of the major characteristics of OODBs is class hierarchy where a subclass is allowed to inherit the definitions defined on its superclasses. In this paper, I present the evaluation metrics for class hierarchy quality in OODBs. These metrics are developed to determine if a concurrency control scheme can achieve good performance or not on a given class hierarchy. I first discuss the existing concurrency control schemes for OODBs. Then I provide evaluation metrics based on structural information and access frequency information in class hierarchies. In order to discuss significance of the proposed performance metrics, an analytical model is developed. Analysis results show that the performance metrics are important factor in concurrency control performance. I consider both single inheritance and multiple inheritance. The proposed metrics can be used to provide guidelines on how to design class hierarchy of an OODB for maximizing the performance of concurrency control technique.

• 한국멀티미디어학회논문지
• /
• 제23권10호
• /
• pp.1236-1249
• /
• 2020

In a most recent object tracking research work, applying Convolutional Neural Network and Recurrent Neural Network-based strategies become relevant for resolving the noticeable challenges in it, like, occlusion, motion, object, and camera viewpoint variations, changing several targets, lighting variations. In this paper, the LSTM Network-based Tracking association method has proposed where the technique capable of real-time multi-object tracking by creating one of the useful LSTM networks that associated with tracking, which supports the long term tracking along with solving challenges. The LSTM network is a different neural network defined in Keras as a sequence of layers, where the Sequential classes would be a container for these layers. This purposing network structure builds with the integration of tracking association on Keras neural-network library. The tracking process has been associated with the LSTM Network feature learning output and obtained outstanding real-time detection and tracking performance. In this work, the main focus was learning trackable objects locations, appearance, and motion details, then predicting the feature location of objects on boxes according to their initial position. The performance of the joint object tracking system has shown that the LSTM network is more powerful and capable of working on a real-time multi-object tracking process.

• 한국멀티미디어학회논문지
• /
• 제20권6호
• /
• pp.893-910
• /
• 2017

In this paper, we propose a simple online multiple object (human) tracking method, LKDeep (Lucas-Kanade feature and Detection based Simple Online Multiple Object Tracker), which can run in fast online enough on CPU core only with acceptable tracking performance for embedded surveillance purpose. The proposed LKDeep is a pragmatic hybrid approach which tracks multiple objects (humans) mainly based on LK features but is compensated by detection on periodic times or on necessity times. Compared to other state-of-the-art multiple object tracking methods based on 'Tracking-By-Detection (TBD)' approach, the proposed LKDeep is faster since it does not have to detect object on every frame and it utilizes simple association rule, but it shows a good object tracking performance. Through experiments in comparison with other multiple object tracking (MOT) methods using the public DPM detector among online state-of-the-art MOT methods reported in MOT challenge [1], it is shown that the proposed simple online MOT method, LKDeep runs faster but with good tracking performance for surveillance purpose. It is further observed through single object tracking (SOT) visual tracker benchmark experiment [2] that LKDeep with an optimized deep learning detector can run in online fast with comparable tracking performance to other state-of-the-art SOT methods.

• 한국멀티미디어학회논문지
• /
• 제21권12호
• /
• pp.1425-1430
• /
• 2018

In this paper, we propose a CNN based deep learning algorithm for semantic segmentation of images. In order to improve the accuracy of semantic segmentation, we combined pixel level object classification and image level object classification. The image level object classification is used to accurately detect the characteristics of an image, and the pixel level object classification is used to indicate which object area is included in each pixel. The proposed network structure consists of three parts in total. A part for extracting the features of the image, a part for outputting the final result in the resolution size of the original image, and a part for performing the image level object classification. Loss functions exist for image level and pixel level classification, respectively. Image-level object classification uses KL-Divergence and pixel level object classification uses cross-entropy. In addition, it combines the layer of the resolution of the network extracting the features and the network of the resolution to secure the position information of the lost feature and the information of the boundary of the object due to the pooling operation.

• 한국멀티미디어학회논문지
• /
• 제17권8호
• /
• pp.915-922
• /
• 2014

We propose an adaptive zoom motion estimation method where a picture is divided into two areas based on the distance information with a depth camera : the one is object area and the other is background area. In the proposed method, the zoom motion is only applied to the object area except the background area. Further, the block size of motion estimation for the object area is set to smaller than that of background area. This adaptive zoom motion estimation method can be reduced at the complexity of motion estimation and can be improved at the motion estimation performance by reducing the block size of the object area in comparison with the conventional zoom motion estimation method. Based on the simulation results, the proposed method is compared with the conventional methods in terms of motion estimation accuracy and computational complexity.

• 한국멀티미디어학회논문지
• /
• 제18권5호
• /
• pp.575-582
• /
• 2015

Obtaining the complete trajectory of the object is a very important task in computer vision applications, such as video surveillance. Previous studies to recover the trajectory between two disconnected trajectory segments, however, do not takes into account the object's motion characteristics and uncertainty of trajectory segments. In this paper, we present a novel approach to recover the trajectory between two disjoint but associated trajectory segments, called goal-directed tracking. To incorporate the object's motion characteristics and uncertainty, the goal-directed state equation is first introduced. Then the goal-directed tracking framework is constructed by integrating the equation to the object tracking and trajectory linking process pipeline. Evaluation on challenging dataset demonstrates that proposed method can accurately recover the missing trajectory between two disconnected trajectory segments as well as appropriately constrain a motion of the object to the its goal(or the target state) with uncertainty.

• 한국멀티미디어학회논문지
• /
• 제20권8호
• /
• pp.1200-1207
• /
• 2017

Sensors in a smartphone can be used to measure various physical quantities. In this paper, we propose an object localization scheme in a three dimenstional using a smart phone. The proposed scheme estimates the location of an object by observing it from several different points. The direction to the target object and the locations of the observation points are collected at each observation point using the location sensor and the orientation sensor in the smartphone. Based on these observations, the proposed scheme derives three dimensional line of sight vectors and estimates the location of the target object that minimizes the estimation error. We implemented the proposed scheme on an Android smartphone and tested its performance by estimating the height of a building and characteristics of the proposed approach.

• 한국멀티미디어학회논문지
• /
• 제2권2호
• /
• pp.195-199
• /
• 1999

기존의 객체 지향 모델들은 공유와 재사용 측면에서 효율적이지만, 아쉽게도 객체의 사용 범위를 정의할 수 있는 방법이 결여되어 있다. 본 논문에서는 객체의 사용 범위를 정의할 수 있도록 하는 새로운 객체 지향 모델을 정의한다 즉 객체는 필요할 동안만 프로그램 내에 존재하게 된다 따라서 이 모텔에서 현재 사용중인 객체의 집합은 동적으로 변할 수 있고, 필요한 객체들만이 관리된다 우리는 예제들을 통하여 이 모델의 유용성을 보인다.

• 한국멀티미디어학회논문지
• /
• 제24권2호
• /
• pp.167-177
• /
• 2021

In the field of augmented reality, edge-based methods have been popularly used in tracking textureless 3D objects. However, edge-based methods are inherently vulnerable to cluttered backgrounds. Another way to track textureless or poorly-textured 3D objects is to directly align image intensity of 3D object between consecutive frames. Although the direct methods enable more reliable and stable tracking compared to using local features such as edges, they are more sensitive to occlusion and less accurate than the edge-based methods. Therefore, we propose a method that combines an edge-based method and a direct method to leverage the advantages from each approach. Experimental results show that the proposed method is much robust to both fast camera (or object) movements and occlusion while still working in real time at a frame rate of 18 Hz. The tracking success rate and tracking accuracy were improved by up to 84% and 1.4 pixels, respectively, compared to using the edge-based method or the direct method solely.

• 한국멀티미디어학회논문지
• /
• 제24권3호
• /
• pp.345-354
• /
• 2021

Recently, deep learning-based image recognition systems have been adopted to various surveillance environments, but most of them are still picture-type object recognition methods, which are insufficient for the long term temporal analysis and high-dimensional situation management. Therefore, we propose a method recognizing the specific dangerous situation generated by human in real-time, and utilizing deep learning-based object analysis techniques. The proposed method uses deep learning-based object detection and tracking algorithms in order to recognize the situations such as 'trespassing', 'loitering', and so on. In addition, human's joint pose data are extracted and analyzed for the emergent awareness function such as 'falling down' to notify not only in the security but also in the emergency environmental utilizations.