Bong-Nam Kang

The POSTECH face database (PF07) and performance evaluation

We constructed a face database POSTECH face database (PF07). PF07 contains the true-color face images of 200 people, 100 men and 100 women, representing 320 various images (5 pose variations times 4 expression variations times 16 illumination variations) per person. All of the people in the database are Korean. We also present the results of face recognition experiments under various conditions using three baseline face recognition algorithms in order to provide an example evaluation protocol on the database. The database is expected to be used to evaluate the algorithm of face recognition for Korean people or for people with systematic variations.

Hidden Markov Model Ensemble for Activity Recognition Using Tri-Axis Accelerometer

Recently, thanks to a variety of sensors equipped on smartphone, a lot of research about mobile activity recognition using accelerometer have been studied for context inference of mobile user and healthcare applications. Previous works, however, have a limitation in classifying some activities because of intra-class variations and inter-class similarities. To handle this problem, in this paper we propose a novel method to recognize activity of smart phone user based on hidden Markov model, where an ensemble method of hidden Markov models is proposed and used to recognize activity. To evaluate our method, we have carried out some experiments by using UCI Human Activity Recognition dataset, and as a result we have achieved about 83.51% accuracy when using two simple features, mean and standard deviation. It is a comparable result to other powerful discriminative methods such as support vector machine and multilayer perceptron.

Face identification using affine simulated dense local descriptors

In this paper, we propose a method for pose and facial expression invariant face identification using the affine simulated local descriptors. Although the currently studied approaches present the higher recognition rate for face verification, the performance of face identification are still low. The proposed method consist of four step, we first normalize the face image using the face detector and eye detector. In second step, we apply the affine simulation for synthesizing various viewed face images. In third step, we make a descriptor on the overlapping block-based grid keypoints. In final step, a probe image is compared with the reference images in a gallery by calculating the number of nearest neighbor keypoints. To improve the recognition performance, we use also the keypoint distance ratio and false matched keypoint ratio. The proposed method using the affine simulated local descriptors showed the better performance than that of cosine similarity metric learning (CSML) method in terms of true acceptance rate, false rejection rate, false acceptance rate, and Rank-1 recognition rate.

PFW: A Face Database in the Wild for Studying Face Identification and Verification in Uncontrolled Environment

To train and evaluate various face recognition algorithms, quite many databases have been created. But most of them have been created under controlled conditions to study the specific variations of the face recognition problem. These variations include position, pose, lighting, background, camera quality and gender. But in real environment, there are also many applications in which there is little or no control over such variations. Labeled Faces in the Wild, a database has been provided to study the latter, unconstrained face recognition problem. However, LFW is proposed for face verification problem, while we observe that a good verification performance cannot guarantee a good identification performance in real situation. Further, the face images in LFW are not sufficient for training to get a state of the art performance. PFW, POS Faces in the Wild, on the contrast, is a large database which can be served both for evaluating face verification and face identification algorithms. Specifically, PFW contains a certain number of identities and each identity contains quite many images, thus make it suitable both for large scale supervised and semi supervised training. In this paper, we also provide some rules for evaluating the identification algorithm performance in real environment. To the best of our knowledge, our database is the first public available large face data set proposed for face identification in unconstrained environment.

Abnormal Object Detection Using Feedforward Model and Sequential Filters

Abnormal object detection and discrimisnation is a critical research area for vision-based surveillance systems. This paper proposes a novel algorithm for the detection and discrimination of abnormal objects, such as abandoned and stolen objects. The proposed algorithm consists of three stages and three different filters. The three stages cooperate with each other using the feedforward model to enhance detection and discrimination performance, while the sequential filters efficiently reject falsely detected regions using three categories of information. The results of experiments conducted using public datasets indicate that the proposed algorithm is more accurate and has a lower false alarm ratio than the existing system.

Face recognition using affine dense SURF-like descriptors

In this paper, we propose the method for pose and facial expression invariant face recognition using the affine dense SURF-like descriptors. The proposed method consists of four step, 1) we normalize the face image using the face and eye detector. 2) We apply the affine simulation for synthesizing various pose face images. 3) We make a descriptor on the overlapping block-based grid keypoints. 4) A probe image is compared with the referenced images by performing the nearest neighbor matching. To improve the recognition rate, we use the keypoint distance ratio and false matched keypoint ratio. The proposed method showed the better performance than that of the conventional methods in terms of the recognition rates.

Multi-resolution 3D morphable models and its matching method

The inverse compositional image alignment (ICIA) is known as an efficient matching method for 3D morphable models (3DMMs). However, it requires a long computation time since the 3D face models consist of a large number of vertices. Also, it requires to recompute the Hessian matrix using the visible vertices every iteration. For a fast and an efficient matching, we propose the efficient and accurate hierarchical ICIA (HICIA) matching method for 3DMMs. The proposed matching method requires multi-resolution 3D face models and the Gaussian image pyramid. The multi-resolution 3D face models are built by sub-sampling at the 2:1 sampling rate to construct the lower-resolution 3D face models. For more accurate matching, we use a two-stage model parameter update that only updates the rigid and the texture parameters and then updates all parameters after the initial convergence. We present several experimental results to prove that the proposed method shows better performance than that of the conventional ICIA matching method.

Nonparametric state machine with multiple features for abnormal object classification

Abandoned object and removed object are important abnormal objects in visual surveillance area to predict the crimes such as explosion or theft event. In real situations, most of existing methods using CCD camera show inconsistent performance because they use a lot of threshold values depending on the environmental conditions of target scene such as illumination change, high traffic volume and complex background. We propose a nonparametric state machine with hierarchical structure consisting of three layers. As shown in the experimental results, the proposed method can be applied to general situations because the state transitions is performed by trained SVM classifiers.

Car License Plate Detection under Large Variations Using Covariance and HOG Descriptors

This paper presents a novel method that can detect license plates which have large variations including perspective distortion, size variation, blurring. Spatial combinations of covariance descriptors in different positions are used with feed-forward network to extract plate-like region and HOG descriptor is used with LDA for validation. From this method, we could achieve high detection rate 94% while maintaining low FPPW(2.5− 6) in road view image.

Reliable multi-person identification using DCNN-based face recognition algorithm and scale-ratio method

Recently, deep convolutional neural networks (DC-NNs) have set a new trend in the computer vision community by improving the state-of-the-art performance in almost all of applications. We propose DCNN-based face recognition algorithm. This paper aims at analyzing and verifying considerations when the proposed method is implemented in a real environment. First, Multiple images of the same scene are processed. Also, this novel method is considered together with a method to reduce the total recognition time in the perspective of integrated system. We analyzed the experimental data and evaluated the performance by setting up the sensor fusion network in actual classroom.