Showing papers by "Dong Sun Park published in 2016"
TL;DR: A novel single object tracking-by-detection tracker is proposed in this paper, combining semi-supervised learning, multiple instance learning and the Bayesian theorem, which has excellent performance over other state-of-the-art trackers for thirteen open-access video sequences.
Abstract: Adaptive discriminative tracking is a new research topic that has attracted broad attention due to its extensive application value. To take full advantage of the information about targets and their surrounding background, we propose a novel single object tracking-by-detection tracker in this paper, combining semi-supervised learning, multiple instance learning and the Bayesian theorem. The tracker uses a block-based inconsistency function of the labeled and unlabeled training samples in the selection of optimal weak classifiers during the parameter updating phase of each frame. Experimental results showed that the proposed tracker has excellent performance over other eight state-of-the-art trackers for thirteen open-access video sequences.
20 citations
30 Nov 2016
TL;DR: This paper proposes a way to utilize a Faster Region based Convolutional Neural Networks (Faster R-CNN) and a Conventional convolutional neural Networks (CNN), which improves the computational speed and is robust against changed environments.
Abstract: Automatic License Plate Detection (ALPD) is a key technology for a efficient traffic control. It is used to improve work efficiency in many applications such as toll payment systems and parking and traffic management. Until recently, the hand-crafted features made for image processing are used to detect license plates in most studies. It has the advantage in speed. but can degrade the detection rate with respect to various environmental changes. In this paper, we propose a way to utilize a Faster Region based Convolutional Neural Networks (Faster R-CNN) and a Conventional Convolutional Neural Networks (CNN), which improves the computational speed and is robust against changed environments. The module based on Faster R-CNN is used to detect license plate candidate regions from images and is followed by the module based on CNN to remove False Positives from the candidates. As a result, we achieved a detection rate of 99.94% from images captured under various environments. In addition, the average operating speed is 80ms/image. We implemented a fast and robust Real-Time License Plate Detection System.
10 citations
01 Jan 2016
TL;DR: People re-identification( people re-id)은 두 영상의 사 람이 �’ 학습시킨 CNN 특징 추출 방법을 스스로 �’ 습한 것이다.
Abstract: People re-identification(People re-id)은 두 영상의 사 람이 같은 사람인지 판단하는 것이다. 일반적으로 지능 형 감시 시스템에서 두 영상은 서로 중첩되지 않은 카메 라에서 획득한 영상이다. People re-id에서는 빛의 변화, 촬영 방향의 변화, 카메라 파라미터의 변화, 포즈의 변 화, 겹침 현상(Occlusion), 낮은 해상도 등 많은 문제를 격고 있다. 최근 객체인식 분야에서는 Convolutional Neural Network (CNN)이 주목받고 있다. CNN의 특징 중 하나 는 입력이미지로 부터 특징 추출 방법을 스스로 학습한 다는 것이다. 전통적은 객체인식 방법에서는 handwritten feature extractor를 사용하지만, CNN은 스스로가 특징을 추출한다. CNN은 많은 학습 데이터를 필요로 하지만 People re-id 데이터 셋의 학습 데이터는 그리 많지 않다. 이 논 문에서 우리는 학습시킨 CNN의 특징 추출 능력이 다른 응용에 대해서도 효과가 있는지 실험하였다. 객체인식 데이터로 pre-train된 CNN을 사용하여 특징벡터를 추출 하였고, 다른 학습 없이 특징벡터의 거리를 사용하여 people re-id를 하였다. VGG 넷[1]은 Large Scale Visual Recognition Challenge 2014 (ILSVRC2014)의 Classification and localization 부분에서 좋은 성능을 보 였다. 우리가 사용한 VGG-S 넷[2]은 VGG 넷의 버전 중 하나이다. II. 제안하는 시스템