• Proceedings of the KIEE Conference
• /
• 2006.10c
• /
• pp.300-302
• /
• 2006

The structured lighting technique is a very accurate depth measurement method. One problem in this technique is that the laser light is invisible very often partly or as a whole caused from occlusion, light absorbtion, and reflection on the surface. If there is no laser light detected on a row of the image frame, the depth could not be computed at the point on the row. In this paper, a Dynamic Programing-based efficient laser line tracking algorithm is proposed to find the most likely line of the laser light utilizing all the possible information of the laser light trace. The performance of the algorithm is shown.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.714-718
• /
• 2009

3D reconstruction of a human face from an image sequence remains an important problem in computer vision. We propose a method, based on a factorization algorithm, that reconstructs a 3D face model from short image sequences exhibiting rotational motion. Factorization algorithms can recover structure and motion simultaneously from one image sequence, but they usually require that all feature points be well tracked. Under rotational motion, however, feature tracking often fails due to occlusion and frame out of features. Additionally, the paucity of images may make feature tracking more difficult or decrease reconstruction accuracy. The proposed 3D reconstruction approach can handle short image sequences exhibiting rotational motion wherein feature points are likely to be missing. We implement the proposal as a reconstruction method; it employs image sequence division and a feature tracking method that uses Active Appearance Models to avoid the failure of feature tracking. Experiments conducted on an image sequence of a human face demonstrate the effectiveness of the proposed method.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.5
• /
• pp.505-510
• /
• 2009

In recent years the space where a robot works has been expanding to the human space unlike traditional industrial robots that work only at fixed positions apart from humans. A human in the recent situation may be the owner of a robot or the target in a robotic application. This paper deals with the latter case; when a robot vision system is employed to monitor humans for a surveillance application, each person in a scene needs to be identified. Humans, however, often move together, and occlusions between them occur frequently. Although this problem has not been seriously tackled in relevant literature, it brings difficulty into later image analysis steps such as tracking and scene understanding. In this paper, a probabilistic neural network is employed to learn the patterns of the best dividing position along the top pixels of an image region of partly occlude people. As this method uses only shape information from an image, it is simple and can be implemented in real time.

• Journal of Korean Academy of Oral and Maxillofacial Radiology
• /
• v.15 no.1
• /
• pp.7-12
• /
• 1985

Anatomical anomaly of temporo-madiblar joint and its dysfunction is becoming one of the important problem in dentistry because the number of these cases are increasing rapidly. Applying 'Lateral Transcranial Technic', 'Updegrave method' and 'Denar Accuard 100' to skull and adult with normal occlusion the author obtained following results: 1. Grewcock method combied with cephalostat as one of 'Lateral Transcranial Technic' revealed clear picture but the image of condyle head was tend to incline downward. 2. Direction of central radiation in 'Updegrave method' should be 2 recommended inch upward from auditorial mearus. 3. For functional analysis and correct diagnosis 'Denar Accuard 100' should be highly recommended.

• The Journal of the Korean dental association
• /
• v.52 no.8
• /
• pp.491-505
• /
• 2014

Food impaction between the implant prostheses and adjacent teeth is the most frequently observed problem. It may be caused by the migration of the adjacent teeth. This symptom may be observed at the mesial aspect of implant prostheses especially, and related with the multiple contributing factors including teeth vitality and antagonist. Idal proximal contact with optimal strength and shape should be made for preventing the food impaction. Shape of customized abutment and prosthesis should have optimal emergence profile. Long duration from the extraction to the delivery of implant prostheses, the adjacent teeth and antagonist teeth may have possibility of occlusal interferences. Remained teeth mobility can induce the food impaction regardless of interproximal contact strength. Occlusal adjustment to remove occlusal interferences can be a method for enhancing the stability of interproximal contact.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10b
• /
• pp.616-618
• /
• 2003

In this paper particle filter is used as an underlying algorithm to track multiple objects, which are soccer players. Multi-object tracking becomes difficult when two or more players get close to and overlap each other because particles of the filters tend to move to a region of higher posterior probability. To resolve this problem, a blob assignation algorithm which identifies the separated image blobs after occlusion, based on the predicted states according to the dynamic model is suggested. This method performed well on the sequences under general camera work.

• Journal of Information Processing Systems
• /
• v.13 no.4
• /
• pp.818-833
• /
• 2017

This paper introduces a new algorithm that renders motion blur using triangular motion paths. A triangle occupies a set of pixels when moving from a position in the start of a frame to another position in the end of a frame. This is a motion path of a moving triangle. For a given pixel, we use a motion path of each moving triangle to find a range of time that this moving triangle is visible to the camera. Then, we sort visible time ranges in the depth-time dimensions and use bitwise operations to solve the occlusion problem. Thereafter, we compute an average color of each moving triangle based on its visible time range. Finally, we accumulate an average color of each moving triangle in the front-to-back order to produce the final pixel color. Thus, our algorithm performs shading after the visibility test and renders motion blur in real time.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.472-474
• /
• 2019

Because of the difference in network structure and loss function, Verification and identification models have their respective advantages and limitations for person reidentification (re-ID). In this work, we propose a multi-task network simultaneously computes the identification loss and verification loss for person reidentification. Given a pair of images as network input, the multi-task network simultaneously outputs the identities of the two images and whether the images belong to the same identity. In experiments, we analyze the major factors affect the accuracy of person reidentification. To address the occlusion problem and improve the generalization ability of reID models, we use the Random Erasing Augmentation (REA) method to preprocess the images. The method can be easily applied to different pre-trained networks, such as ResNet and VGG. The experimental results on the Market1501 datasets show significant and consistent improvements over the state-of-the-art methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.28.2-28
• /
• 2001

Three dimensional shape measurement techniques are widely needed in industries for product quality monitoring and control. X-ray imaging method is a promising technology to achieve three-dimensional Information, both the surface and inner structure of an object, since it can overcome the limitations of conventional visual or optical methods such as an occlusion problem or surface reflection properties. In this paper, we propose three dimensional volume reconstruction method based on x-ray stereo imaging technology. Here, the stereo images of an object from two different views are taken by changing the object pose rather than moving imaging plane as in conventional stereo vision method. We propose a series of image processing techniques to extract the features efficiently from x-ray images, where the occluded features in case of normal camera vision could be found ...

• Journal of Oral Medicine and Pain
• /
• v.46 no.4
• /
• pp.155-160
• /
• 2021

The masticatory muscle disorder is the most common problem that patients with temporomandibular disorder often complain. For such complaints, treatment is directed towards reducing hyperactivity of muscles or effects of the central nervous system. However, if nonspecific occlusal change or pain persists, it is necessary to consider that muscle weakness might be the cause of the persistence of temporomandibular disorder. Stabilization of occlusion and improvement of the pain symptoms were achieved in both cases through the chewing gum exercise. This exercise may enable masticatory movements done in normal function by using muscle engram and achieve reinforcement of the masticatory muscles with balanced, simultaneous contacts of the teeth. In addition, it may be a viable method for treating temporomandibular disorders that do not respond well to conventional mandibular stabilization therapies.