• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.12
• /
• pp.6009-6027
• /
• 2019

Support vector machine (SVM) classifiers have been widely used for object detection. These methods usually locate the object by finding the region with maximal score in an image. With bag-of-features representation, the SVM score of an image region can be written as the sum of its inside feature-weights. As a result, the searching process can be executed efficiently by using strategies such as branch-and-bound. However, the feature-weight derived by optimizing region classification cannot really reveal the category knowledge of a feature-point, which could cause bad localization. In this paper, we represent a region in an image by a collection of local feature-points and determine the object by the region with the maximum posterior probability of belonging to the object class. Based on the Bayes' theorem and Naive-Bayes assumptions, the posterior probability is reformulated as the sum of feature-scores. The feature-score is manifested in the form of the logarithm of a probability ratio. Instead of estimating the numerator and denominator probabilities separately, we readily employ the density ratio estimation techniques directly, and overcome the above limitation. Experiments on a car dataset and PASCAL VOC 2007 dataset validated the effectiveness of our method compared to the baselines. In addition, the performance can be further improved by taking advantage of the recently developed deep convolutional neural network features.

• Journal of Korea Multimedia Society
• /
• v.11 no.12
• /
• pp.1658-1667
• /
• 2008

Reconstruction of 3D objects from a single view image is generally an ill-posed problem because of the projection distortion. A monocular vision based 3D object localization method is proposed in this paper, which approximates an object on the ground to a simple bounding solid and works automatically without any prior information about the object. A spherical or cylindrical object determined based on a circularity measure is approximated to a bounding cylinder, while the other general free-shaped objects to a bounding box or a bounding cylinder appropriately. For a general object, its silhouette on the ground is first computed by back-projecting its projected image in image plane onto the ground plane and then a base rectangle on the ground is determined by using the intuition that touched parts of the object on the ground should appear at lower part of the silhouette. The base rectangle is adjusted and extended until a derived bounding box from it can enclose the general object sufficiently. Height of the bounding box is also determined enough to enclose the general object. When the general object looks like a round-shaped object, a bounding cylinder that encloses the bounding box minimally is selected instead of the bounding box. A bounding solid can be utilized to localize a 3D object on the ground and to roughly estimate its volume. Usefulness of our approach is presented with experimental results on real image objects and limitations of our approach are discussed.

• The Magazine of the IEIE
• /
• v.39 no.2
• /
• pp.45-54
• /
• 2012

Object detection is a fundamental task for many high-level computer vision applications such as image retrieval, scene understanding, activity recognition, visual surveillance and many others. Although object detection is one of the most popular problems in computer vision and various algorithms have been proposed thus far, it is also notoriously difficult, mainly due to lack of proper models for object representation, that handle large variations of object structure and appearance. In this article, we review a branch of object detection algorithms based on Support Vector Machines (SVMs), a well-known max-margin technique to minimize classification error. We introduce a few variations of SVMs-Structural SVMs and Latent SVMs-and discuss their applications to object detection and localization.

• Journal of Internet Computing and Services
• /
• v.18 no.6
• /
• pp.85-92
• /
• 2017

Recently, in the field of video surveillance, a Deep Learning based learning method has been applied to a method of detecting a moving person in a video and analyzing the behavior of a detected person. The human activity recognition, which is one of the fields this intelligent image analysis technology, detects the object and goes through the process of detecting the body keypoint to recognize the behavior of the detected object. In this paper, we propose a method for Body Keypoint Localization based on Object Detection using RGB-D information. First, the moving object is segmented and detected from the background using color information and depth information generated by the two cameras. The input image generated by rescaling the detected object region using RGB-D information is applied to Convolutional Pose Machines for one person's pose estimation. CPM are used to generate Belief Maps for 14 body parts per person and to detect body keypoints based on Belief Maps. This method provides an accurate region for objects to detect keypoints an can be extended from single Body Keypoint Localization to multiple Body Keypoint Localization through the integration of individual Body Keypoint Localization. In the future, it is possible to generate a model for human pose estimation using the detected keypoints and contribute to the field of human activity recognition.

• Smart Media Journal
• /
• v.12 no.9
• /
• pp.45-59
• /
• 2023

Weakly supervised object localization (WSOL) is a task of localizing an object in an image using only image-level labels. Previous studies have followed the conventional class activation mapping (CAM) pipeline. However, we reveal the current CAM approach suffers from problems which cause original CAM could not capture the complete defects features. This work utilizes a convolutional neural network (CNN) pretrained on image-level labels to generate class activation maps in a multi-scale manner to highlight discriminative regions. Additionally, a vision transformer (ViT) pretrained was treated to produce multi-head attention maps as an auxiliary detector. By integrating the CNN-based CAMs and attention maps, our approach localizes defective regions without requiring bounding box or pixel-level supervision during training. We evaluate our approach on a dataset of apple images with only image-level labels of defect categories. Experiments demonstrate our proposed method aligns with several Object Detection models performance, hold a promise for improving localization.

• The Journal of Korea Robotics Society
• /
• v.9 no.4
• /
• pp.232-241
• /
• 2014

This paper proposes an underwater localization algorithm using probabilistic object recognition. It is organized as follows; 1) recognizing artificial objects using imaging sonar, and 2) localizing the recognized objects and the vehicle using EKF(Extended Kalman Filter) based SLAM. For this purpose, we develop artificial landmarks to be recognized even under the unstable sonar images induced by noise. Moreover, a probabilistic recognition framework is proposed. In this way, the distance and bearing of the recognized artificial landmarks are acquired to perform the localization of the underwater vehicle. Using the recognized objects, EKF-based SLAM is carried out and results in a path of the underwater vehicle and the location of landmarks. The proposed localization algorithm is verified by experiments in a basin.

• The Journal of Korea Robotics Society
• /
• v.1 no.2
• /
• pp.117-124
• /
• 2006

In this paper, we develope the navigation system for patrol robots in indoor environment. The proposed system consists of PDA map modelling, a localization algorithm based on a global position sensor and an automatic charging station. For the practical use in security system, the PDA is used to build object map on the given indoor map. And the builded map is downloaded to the mobile robot and used in path planning. The global path planning is performed with a localization sensor and the downloaded map. As a main controller, we use PXA270 based hardware platform in which embedded linux 2.6 is developed. Data handling for various sensors and the localization algorithm are performed in the linux platform. Also, we implemented a local path planning algorithm for object avoidance with ultra sonar sensors. Finally, for the automatic charging, we use an infrared ray system and develop a docking algorithm. The navigation system is experimented with the two-wheeled mobile robot using North-Star localization system.

• The Journal of Korea Robotics Society
• /
• v.15 no.4
• /
• pp.357-364
• /
• 2020

Indoor localization using an artificial marker plays a key role for a robot to be used in a service environment. A number of researchers have predefined the positions of markers and attached them to the positions in order to reduce the error of the localization method. However, it is practically impossible to attach a marker to the predetermined position accurately. In order to visualize the position of an object in the environment based on the marker attached to them, it is necessary to consider a change of marker's position or the addition of a marker because of moving the existed object or adding a new object. In this paper, we studied the method to estimate the artificial marker's global position for the visualization of environment. The system calculates the relative distance from a reference marker to others repeatedly to estimate the marker's position. When the marker's position is changed or new markers are added, our system can recognize the changed situation of the markers. To verify the proposed system, we attached 12 markers at regular intervals on the ceiling and compared the estimation result of the proposed method and the actual distance. In addition, we compared the estimation result when changing the position of an existing marker or adding a new marker.

• Journal of Institute of Control, Robotics and Systems
• /
• v.11 no.5
• /
• pp.398-405
• /
• 2005

The intelligent robots that will be needed in the near future are human-friendly robots that are able to coexist with humans and support humans effectively. To realize this, robots need to recognize their position and posture in known environment as well as unknown environment. Moreover, it is necessary for their localization to occur naturally. It is desirable for a robot to estimate of his position by solving uncertainty for mobile robot navigation, as one of the best important problems. In this paper, we describe a method for the localization of a mobile robot using image information of a moving object. This method combines the observed position from dead-reckoning sensors and the estimated position from the images captured by a fixed camera to localize a mobile robot. Using a priori known path of a moving object in the world coordinates and a perspective camera model, we derive the geometric constraint equations which represent the relation between image frame coordinates for a moving object and the estimated robot's position. Also, the control method is proposed to estimate position and direction between the walking human and the mobile robot, and the Kalman filter scheme is used for the estimation of the mobile robot localization. And its performance is verified by the computer simulation and the experiment.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.11
• /
• pp.933-938
• /
• 2001

In this paper, we describe a method for the mobile robot using images of a moving object. This method combines the observed position from dead-reckoning sensors and the estimated position from the images captured by a fixed camera to localize a mobile robot. Using the a priori known path of a moving object in the world coordinates and a perspective camera model, we derive the geometric constraint equations which represent the relation between image frame coordinates for a moving object and the estimated robot`s position. Since the equations are based on the estimated position, the measurement error may exist all the time. The proposed method utilizes the error between the observed and estimated image coordinates to localize the mobile robot. The Kalman filter scheme is applied to this method. Effectiveness of the proposed method is demonstrated by the simulation.