• 제목/요약/키워드: Occlusions

검색결과 185건 처리시간 0.021초

Enhancing Occlusion Robustness for Vision-based Construction Worker Detection Using Data Augmentation

  • Kim, Yoojun;Kim, Hyunjun;Sim, Sunghan;Ham, Youngjib
    • 국제학술발표논문집
    • /
    • The 9th International Conference on Construction Engineering and Project Management
    • /
    • pp.904-911
    • /
    • 2022
  • Occlusion is one of the most challenging problems for computer vision-based construction monitoring. Due to the intrinsic dynamics of construction scenes, vision-based technologies inevitably suffer from occlusions. Previous researchers have proposed the occlusion handling methods by leveraging the prior information from the sequential images. However, these methods cannot be employed for construction object detection in non-sequential images. As an alternative occlusion handling method, this study proposes a data augmentation-based framework that can enhance the detection performance under occlusions. The proposed approach is specially designed for rebar occlusions, the distinctive type of occlusions frequently happen during construction worker detection. In the proposed method, the artificial rebars are synthetically generated to emulate possible rebar occlusions in construction sites. In this regard, the proposed method enables the model to train a variety of occluded images, thereby improving the detection performance without requiring sequential information. The effectiveness of the proposed method is validated by showing that the proposed method outperforms the baseline model without augmentation. The outcomes demonstrate the great potential of the data augmentation techniques for occlusion handling that can be readily applied to typical object detectors without changing their model architecture.

  • PDF

Robust Multithreaded Object Tracker through Occlusions for Spatial Augmented Reality

  • Lee, Ahyun;Jang, Insung
    • ETRI Journal
    • /
    • 제40권2호
    • /
    • pp.246-256
    • /
    • 2018
  • A spatial augmented reality (SAR) system enables a virtual image to be projected onto the surface of a real-world object and the user to intuitively control the image using a tangible interface. However, occlusions frequently occur, such as a sudden change in the lighting environment or the generation of obstacles. We propose a robust object tracker based on a multithreaded system, which can track an object robustly through occlusions. Our multithreaded tracker is divided into two threads: the detection thread detects distinctive features in a frame-to-frame manner, and the tracking thread tracks features periodically using an optical-flow-based tracking method. Consequently, although the speed of the detection thread is considerably slow, we achieve real-time performance owing to the multithreaded configuration. Moreover, the proposed outlier filtering automatically updates a random sample consensus distance threshold for eliminating outliers according to environmental changes. Experimental results show that our approach tracks an object robustly in real-time in an SAR environment where there are frequent occlusions occurring from augmented projection images.

Deep Facade Parsing with Occlusions

  • Ma, Wenguang;Ma, Wei;Xu, Shibiao
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제16권2호
    • /
    • pp.524-543
    • /
    • 2022
  • Correct facade image parsing is essential to the semantic understanding of outdoor scenes. Unfortunately, there are often various occlusions in front of buildings, which fails many existing methods. In this paper, we propose an end-to-end deep network for facade parsing with occlusions. The network learns to decompose an input image into visible and invisible parts by occlusion reasoning. Then, a context aggregation module is proposed to collect nonlocal cues for semantic segmentation of the visible part. In addition, considering the regularity of man-made buildings, a repetitive pattern completion branch is designed to infer the contents in the invisible regions by referring to the visible part. Finally, the parsing map of the input facade image is generated by fusing the results of the visible and invisible results. Experiments on both synthetic and real datasets demonstrate that the proposed method outperforms state-of-the-art methods in parsing facades with occlusions. Moreover, we applied our method in applications of image inpainting and 3D semantic modeling.

1급부정교합(一級不正咬合)에 관(關)한 두부방사선계측학적(頭部放射線計測學的) 연구(硏究) (A ROENTGENOCEPHALOMETRIC DESCRIPTION OF CLASS I MALOCCLUSION)

  • 황선문
    • 대한치과교정학회지
    • /
    • 제9권1호
    • /
    • pp.105-110
    • /
    • 1979
  • In order to define what the average Class I malocclusion looks like, 72 Class I malocclusions were compared with corresponding measurements from 90 cases of normal occlusion sample. This sample was obtained from the Department of Orthodontics, Infirmary of Seoul National University. 1. SNA, Fac(NP) to SN measurements were significantly different from the normal occlusions and smaller than the normal occlusion means. 2. 'Y' axis to SN, Mandibular plane to SN, $\underline{1}$ to Occlusal plane, $\bar{1}$ to NB(linear) measurements were significantly different from the normal occlusions and larger than the normal occlusions. 3. ANB, Occlusal plane to SN, AB to Occlusal plane, $\underline{1}$ to SN, $\bar{1}$ to SN, $\bar{1}$ to Occlusal plane measurements showed no significant difference between the means.

  • PDF

Face recognition invariant to partial occlusions

  • Aisha, Azeem;Muhammad, Sharif;Hussain, Shah Jamal;Mudassar, Raza
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제8권7호
    • /
    • pp.2496-2511
    • /
    • 2014
  • Face recognition is considered a complex biometrics in the field of image processing mainly due to the constraints imposed by variation in the appearance of facial images. These variations in appearance are affected by differences in expressions and/or occlusions (sunglasses, scarf etc.). This paper discusses incremental Kernel Fisher Discriminate Analysis on sub-classes for dealing with partial occlusions and variant expressions. This framework focuses on the division of classes into fixed size sub-classes for effective feature extraction. For this purpose, it modifies the traditional Linear Discriminant Analysis into incremental approach in the kernel space. Experiments are performed on AR, ORL, Yale B and MIT-CBCL face databases. The results show a significant improvement in face recognition.

A Newly Designed Nitinol Stent: Early Clinical Experience in the Treatment of Iliac Artery Stenoses and Occlusions

  • Chang Jin Yoon;Jin Wook Chung;Jae Hyung Park;Soon Hyung Hong;Soon Young Song;Hyung Gehn Lim;Yoon Sin Lee
    • Korean Journal of Radiology
    • /
    • 제2권3호
    • /
    • pp.145-150
    • /
    • 2001
  • Objective: To investigate the effectiveness of the newly designed Niti-S stent in the management of iliac arterial stenoses and occlusions. Materials and Methods: Stenoses (n=25) or occlusions (n=5) in the iliac arteries of 25 patients (30 limbs) were treated. The site of the lesions was the common (n=15) or external (n=11) iliac artery, or both (n=4). Eight limbs were treated for diffuse disease, six for highly eccentric lesion, five for occlusive lesion, and 11 for failed angioplasty. Results: In all patients, technical success was achieved without major complications. One death, not procedure-related, occurred within 30 days. Ankle-brachial indexes improved from 0.63 0.30 to 0.99 0.21, and ischemic symptoms showed improvement in 22 patients (88%). Fontaine classifications before stenting, namely IIa(n=3), IIb(n=16), III(n=2), and IV(n=4) improved to I(n=17), IIa(n=5), and IV(n=3). Follow-up over a 27 (mean, 19.8 8)-month period showed that cumulative patency rates were 95.8% at 1 year and 86.2% at 2 and 3 years. No significant decrease in the mean ankle-brachial index was observed. Conclusion: The Niti-S stent appears to be a safe and effective device for the treatment of iliac stenoses and occlusions. These preliminary results require confirmation with a larger series.

  • PDF

편측성순구개열자의 하악골성장에 관한 연구 (A STUDY ON THE MANDIBULAR GROWTH IN SURGICALLY REPAIRED UNILATERAL CLEFT LIP AND PALATE)

  • 박춘근;양원식
    • 대한치과교정학회지
    • /
    • 제19권1호
    • /
    • pp.153-167
    • /
    • 1989
  • This study was undertaken to analyze the growth of mandible in surgically repaired unilateral cleft lip and palate. The subjects consisted of 63 unilateral cleft lip and palate individuals, 60 class III malocclusions and 60 normal occlusions ranging from 6 to 15 years old. Each group was divided into two age groups. (6-10 Y and 11-15 Y) The results obtained from UCLP compared with other groups were as follows: 1. The anteroposterior position of the chin was similar to that of the normal occlusions. 2. The shape of the mandible was similar to that of the class III malocclusions. 3. In mandibular size, ramus height was the smallest among three groups, but body length and overall mandibular length were similar to those of the normal occlusions. 4. The lower border of the mandible was the steepest among three groups and strong vertical or clockwise growth tendency was indicated. 5. The position of condyle in relation to the cranial base showed little difference in three groups. 6. In older age group , vertical growth tendency of the mandible decreased more or less.

  • PDF

Occluded Object Motion Estimation System based on Particle Filter with 3D Reconstruction

  • Ko, Kwang-Eun;Park, Jun-Heong;Park, Seung-Min;Kim, Jun-Yeup;Sim, Kwee-Bo
    • International Journal of Fuzzy Logic and Intelligent Systems
    • /
    • 제12권1호
    • /
    • pp.60-65
    • /
    • 2012
  • This paper presents a method for occluded object based motion estimation and tracking system in dynamic image sequences using particle filter with 3D reconstruction. A unique characteristic of this study is its ability to cope with partial occlusion based continuous motion estimation using particle filter inspired from the mirror neuron system in human brain. To update a prior knowledge about the shape or motion of objects, firstly, fundamental 3D reconstruction based occlusion tracing method is applied and object landmarks are determined. And optical flow based motion vector is estimated from the movement of the landmarks. When arbitrary partial occlusions are occurred, the continuous motion of the hidden parts of object can be estimated by particle filter with optical flow. The resistance of the resulting estimation to partial occlusions enables the more accurate detection and handling of more severe occlusions.

DEM을 이용한 실영상기반 가상표적의 폐색처리기법 (Resolving Occlusion Technique of Virtual Target on Real Image using DEM)

  • 차정희;장효종;김계영
    • 정보처리학회논문지B
    • /
    • 제13B권7호
    • /
    • pp.663-670
    • /
    • 2006
  • 실 세계 영상에 가상표적을 효과적으로 전시하여 현실감을 높이려면 먼저 두 세계를 정합한 후 폐색영역을 산출하여 가상객체의 위치를 결정하는 것이 필수적이다. 본 논문에서는 실 영상위에 지정된 경로에 따라 가상표적을 이동시킬 때 발생하는 폐색문제를 해결하는 새로운 방법을 제안한다. 이를 위해 먼저 실험 영역의 DEM을 이용하여 3차원 가상세계를 생성하고 이를 CCD 카메라 영상과 시각적 단서를 이용하여 정합한다. 또한 스네이크 알고리즘과 픽킹 알고리즘을 이용하여 영상에서 폐색 처리될 지점의 3차원 정보를 산출하고 표적이동시 이를 이용하여 폐색문제를 해결하는 방법을 제안하였다 실험에서는 부분적 폐색이 발생하는 환경에서 제안한 방법의 유효성을 입증하였다.