• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.10
• /
• pp.4080-4097
• /
• 2020

In recent years, convolutional neural network (CNN) has become the primary method for face detection. But its shortcomings are obvious, such as expensive calculation, heavy model, etc. This makes CNN difficult to use on the mobile devices which have limited computing and storage capabilities. Therefore, the design of lightweight CNN for face detection is becoming more and more important with the popularity of smartphones and mobile Internet. Based on the CPU real-time face detector FaceBoxes, we propose a multi-task lightweight face detector, which has low computing cost and higher detection precision. First, to improve the detection capability, the squeeze and excitation modules are used to extract attention between channels. Then, the textual and semantic information are extracted by shallow networks and deep networks respectively to get rich features. Finally, the landmark detection module is used to improve the detection performance for small faces and provide landmark data for face alignment. Experiments on AFW, FDDB, PASCAL, and WIDER FACE datasets show that our algorithm has achieved significant improvement in the mean average precision. Especially, on the WIDER FACE hard validation set, our algorithm outperforms the mean average precision of FaceBoxes by 7.2%. For VGA-resolution images, the running speed of our algorithm can reach 23FPS on a CPU device.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37B no.9
• /
• pp.849-858
• /
• 2012

We proposed a scheme to accurately detect elephant flows. Along the ever increasing traffic trend, certain flows occupy the network heavily in terms of time and network bandwidth. These flows are called elephant flows. Elephant flows raises complicated issues to manage for Internet traffics and services. One of the methods to identify elephant flows is the Landmark LRU cache scheme, which improved the previous method of Least Recently Used scheme. We proposed a cache update algorithm, to further improve the existing Landmark LRU. The proposed scheme improves the accuracy to detect elephant flow while maintaining efficiency of Landmark LRU. We verified our algorithm by simulating on Sangmyung University's wireless real network traces and evaluated the improvement.

• Journal of Broadcast Engineering
• /
• v.26 no.6
• /
• pp.748-757
• /
• 2021

In this paper, we propose a gaze estimation network in which eye landmark position detection and gaze direction vector estimation are integrated into one deep learning network. The proposed network uses the Stacked Hourglass Network as a backbone structure and is largely composed of three parts: a landmark detector, a feature map extractor, and a gaze direction estimator. The landmark detector estimates the coordinates of 50 eye landmarks, and the feature map extractor generates a feature map of the eye image for estimating the gaze direction. And the gaze direction estimator estimates the final gaze direction vector by combining each output result. The proposed network was trained using virtual synthetic eye images and landmark coordinate data generated through the UnityEyes dataset, and the MPIIGaze dataset consisting of real human eye images was used for performance evaluation. Through the experiment, the gaze estimation error showed a performance of 3.9, and the estimation speed of the network was 42 FPS (Frames per second).

• Journal of Korea Multimedia Society
• /
• v.24 no.9
• /
• pp.1242-1250
• /
• 2021

Detection of outliers deviating normal data distribution in high dimensional data is an important technique in many application areas. In this paper, a distance-based outlier detection method using landmarks in high dimensional data is proposed. Given normal training data, the k-means clustering method is applied for the training data in order to extract the centers of the clusters as landmarks which represent normal data distribution. For a test data sample, the distance to the nearest landmark gives the outlier score. In the experiments using high dimensional data such as images and documents, it was shown that the proposed method based on the landmarks of one-tenth of training data can give the comparable outlier detection performance while reducing the time complexity greatly in the testing stage.

• Korean Journal of Remote Sensing
• /
• v.21 no.6
• /
• pp.505-516
• /
• 2005

The Communications, Oceanography and Meteorology Satellite(COMS) will be launched in 2008. Ground processing for COMS includes the process of automatic image navigation. Image navigation requires landmark detection by matching COMS images against landmark chips. For automatic image navigation, a matching must be performed automatically However, if matching results contain errors, the accuracy of Image navigation deteriorates. To overcome this problem, we propose use of a robust estimation technique called Random Sample Consensus (RANSAC) to automatically detect erroneous matching. We tested GOES-9 satellite images with 30 landmark chips that were extracted from the world shoreline database. After matching, mismatch results were detected automatically by RANSAC. All mismatches were detected correctly by RANSAC with a threshold value of 2.5 pixels.

• Proceedings of the KIEE Conference
• /
• 2001.11c
• /
• pp.91-94
• /
• 2001

In this paper, we proposed localization method of mobile robot using color landmark mounted on ceiling. This work is composed 2 parts : landmark recognition part which finds the position of multiple landmarks in image and identifies them and absolute position estimation part which estimates the location and orientation of mobile robot in indoor environment. In landmark recognition part, mobile robot detects artificial color landmarks using simple histogram intersection method in rg color space which is insensitive to the change of illumination. Then absolute position estimation part calculates relative position of the mobile robot to the detected landmarks. For the verification of proposed algorithm, ceiling-orientated camera was installed on a mobile robot and performance of localization was examined by designed artificial color landmarks. As the result of test, mobile robot could achieve the reliable landmark detection and accurately estimate the position of mobile robot in indoor environment.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.9 no.1
• /
• pp.64-69
• /
• 2008

In order that a mobile robot can perform tasks in unknown environment localization of a mobile robot is essential task. In this paper, a new localization method for a mobile robot using an active landmark is proposed, which is very simple to implement. The landmark has a LED which can be controlled by a mobile robot via wireless communication. CCD camera gets two images of the landmark, one of which is with LED off and the other is with LED on. Because the landmark can be detected by using the difference image of the two images, detection time can be minimized. By using the characteristic points of the landmark, localization can be performed simply. A series of experiments are performed to evaluate the proposed method and the experimental results show that the proposed method can be applicable to the localization of a mobile robot.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2003.06a
• /
• pp.306-311
• /
• 2003

This paper presents a novel localization paradigm for mobile robots based on artificial and natural landmarks. A model-based object recognition method detects natural landmarks and conducts the global and topological localization. In addition, a metric localization method using artificial landmarks is fused to complement the deficiency of topology map and guide to action behavior. The recognition algorithm uses a modified local Zernike moments and a probabilistic voting method for the robust detection of objects in cluttered indoor environments. An artificial landmark is designed to have a three-dimensional multi-colored structure and the projection distortion of the structure encodes the distance and viewing direction of the robot. We demonstrate the feasibility of the proposed system through real world experiments using a mobile robot, KASIRI-III.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.9
• /
• pp.749-757
• /
• 2001

We present a simple artificial landmark model and robust landmark tracking algorithm for mobile robot localization. The landmark model, consisting of symmetric and repetitive color patches, produces color histograms that are invariant under the geometric and photometric distortions. A stochastic approach based on the CONDENSATION tracks the landmark model robustly even under the varying illumination conditions. After the landmark detection, relative position of the mobile robot to the landmark is calculated. Experimental results show that the proposed landmark model is effective and can be detected and tracked in a clustered scene robustly. With the tracked single landmark, we extract geometrical information than achieve accurate localization.

• Journal of International Society for Simulation Surgery
• /
• v.2 no.1
• /
• pp.26-32
• /
• 2015

Purpose When a surgeon examines the morphology of skull of patient, locations of craniometric landmarks of 3D computed tomography(CT) volume are one of the most important information for surgical purpose. The locations of craniometric landmarks can be found manually by surgeon from the 3D rendered volume or 2D sagittal, axial, and coronal slices which are taken by CT. Since there are many landmarks on the skull, finding these manually is time-consuming, exhaustive, and occasionally inexact. These inefficiencies raise a demand for a automatic localization technique for craniometric landmark points. So in this paper, we propose a novel method through which we can automatically find these landmark points, which are useful for surgical purpose. Materials and Methods At first, we align the experimental data (CT volumes) using Frankfurt Horizontal Plane (FHP) and Mid Sagittal Plane(MSP) which are defined by 3 and 2 cranial landmark points each. The target landmark of our experiment is the anterior nasal spine. Prior to constructing a statistical cubic model which would be used for detecting the location of the landmark from a given CT volume, reference points for the anterior nasal spine were manually chosen by a surgeon from several CT volume sets. The statistical cubic model is constructed by calculating weighted intensity means of these CT sets around the reference points. By finding the location where similarity function (squared difference function) has the minimal value with this model, the location of the landmark can be found from any given CT volume. Results In this paper, we used 5 CT volumes to construct the statistical cubic model. The 20 CT volumes including the volumes, which were used to construct the model, were used for testing. The range of age of subjects is up to 2 years (24 months) old. The found points of each data are almost close to the reference point which were manually chosen by surgeon. Also it has been seen that the similarity function always has the global minimum at the detection point. Conclusion Through the experiment, we have seen the proposed method shows the outstanding performance in searching the landmark point. This algorithm would make surgeons efficiently work with morphological informations of skull. We also expect the potential of our algorithm for searching the anatomic landmarks not only cranial landmarks.