• Journal of Broadcast Engineering
• /
• v.24 no.6
• /
• pp.1053-1063
• /
• 2019

Recently, tile-based streaming that transmits one 360 video in several tiles, is actively being studied in order to transmit these 360 video more efficiently. In this paper, for the transmission of high-definition 360 video corresponding to user's viewport in tile-based streaming scenarios, a system of assigning the quality of tiles at each tile by applying the saliency map generated by existing network models is proposed. As a result of usage of Motion-Constrained Tile Set (MCTS) technique to encode each tile independently, the user's viewport was rendered and tested based on Salient360! dataset, streaming 360 video based on the proposed system results in gain to 23% of the user's viewport compared to using the existing high-efficiency video coding (HEVC).

• Smart Media Journal
• /
• v.12 no.2
• /
• pp.9-14
• /
• 2023

In this paper, we proposed a face recognition network which attempts to use more facial features awhile using smaller number of training sets. When combining the neural network together for face recognition, we want to use networks that use different part of the facial features. However, the network training chooses randomly where these facial features are obtained. Other hand, the judgment basis of the network model can be expressed as a saliency map through gradCAM. Therefore, in this paper, we use gradCAM to visualize where the trained face recognition model has made a observations and recognition judgments. Thus, the network combination can be constructed based on the different facial features used. Using this approach, we trained a network for small face recognition problem. In an simple toy face recognition example, the recognition network used in this paper improves the accuracy by 1.79% and reduces the equal error rate (EER) by 0.01788 compared to the conventional approach.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.9
• /
• pp.113-127
• /
• 2012

Face detection methods using image processing have been proposed in many different ways. Generally, the most widely used method for face detection is an Adaboost that is proposed by Viola and Jones. This method uses Haar-like feature for image learning, and the detection performance depends on the learned images. It is well performed to detect face images within a certain distance range, but if the image is far away from the camera, face images become so small that may not detect them with the pre-learned Haar-like feature of the face image. In this paper, we propose the far distance face detection method that combine the Aadaboost of Viola-Jones with a saliency map and user's attention information. Saliency Map is used to select the candidate face images in the input image, face images are finally detected among the candidated regions using the Adaboost with Haar-like feature learned in advance. And the user's eye-tracking information is used to select the interest regions. When a subject is so far away from the camera that it is difficult to detect the face image, we expand the small eye gaze spot region using linear interpolation method and reuse that as input image and can increase the face image detection performance. We confirmed the proposed model has better results than the conventional Adaboost in terms of face image detection performance and computational time.

• Journal of Information Processing Systems
• /
• v.13 no.3
• /
• pp.448-463
• /
• 2017

Multimedia is a ubiquitous and indispensable part of our daily life and learning such as audio, image, and video. Objective and subjective quality evaluations play an important role in various multimedia applications. Blind image quality assessment (BIQA) is used to indicate the perceptual quality of a distorted image, while its reference image is not considered and used. Blur is one of the common image distortions. In this paper, we propose a novel BIQA index for Gaussian blur distortion based on the fact that images with different blur degree will have different changes through the same blur. We describe this discrimination from three aspects: color, edge, and structure. For color, we adopt color histogram; for edge, we use edge intensity map, and saliency map is used as the weighting function to be consistent with human visual system (HVS); for structure, we use structure tensor and structural similarity (SSIM) index. Numerous experiments based on four benchmark databases show that our proposed index is highly consistent with the subjective quality assessment.

• International Journal of Contents
• /
• v.10 no.4
• /
• pp.1-10
• /
• 2014

In this paper, we propose a method to build an accurate initial trimap for the GrabCut algorithm without the need for human interaction. First, we identify a rough candidate for the label region of a bottle by applying a saliency map to find a salient area from the image. Then, the Hough Transformation method is used to detect the left and right borders of the label region, and the k-means algorithm is used to localize the upper and lower borders of the label of the bottle. These four borders are used to build an initial trimap for the GrabCut method. Finally, GrabCut segments accurate regions for the label. The experimental results for 130 wine bottle images demonstrated that the saliency map extracted a rough label region with an accuracy of 97.69% while also removing the complex background. The Hough transform and projection method accurately drew the outline of the label from the saliency area, and then the outline was used to build an initial trimap for GrabCut. Finally, the GrabCut algorithm successfully segmented the bottle label with an average accuracy of 92.31%. Therefore, we believe that our method is suitable for product label recognition systems that automatically segment product labels. Although our method achieved encouraging results, it has some limitations in that unreliable results are produced under conditions with varying illumination and reflections. Therefore, we are in the process of developing preprocessing algorithms to improve the proposed method to take into account variations in illumination and reflections.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.1579-1580
• /
• 2015

본 논문에서는 3차원 메쉬 모델의 중요 영역을 표현하는 메쉬 돌출맵(mesh saliency map)을 생성하기 위하여 다중 스케일 평균 곡률 (multi-scale mean curvature)을 기반으로 정의된 전역 희소치(global rarity)를 이용하는 방법을 제안한다. 제안 방법에서는 우선, 메쉬 모델의 지역 영역 특성을 정의하기 위하여 기존 관련 연구들에서 많이 사용하고 있는 가우시안 가중치 평균곡률(Gaussian-weighted mean curvature)을 5단계 서로 다른 스케일에서 정의하고, 메쉬의 각 정점(vertex)에 대하여 중심주변 연산자(center-surround operator)를 적용하여 5단계 지역 돌출특성(local saliency)을 정의한다. 주어진 메쉬 모델의 전역 희소치를 구하기 위하여 메쉬의 모든 정점쌍 (vertex pair)에 대하여 5단계 지역 돌출 특성 공간에서의 거리를 계산하고, 각 정점별로 5단계 지역 돌출 특성 공간에서의 다른 정점과의 거리의 합으로 전역 희소치를 정의한다. 이러한 전역 희소치를 각 정점의 메쉬 돌출치로 정의한다. 서로 다른 형태의 3차원 모델에 대하여 제안방법에 의한 메쉬 돌출맵과 지역 특성만을 고려한 기존 메쉬 돌출맵을 생성하여 중요 영역 표현 결과를 비교 분석한다.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.8
• /
• pp.123-132
• /
• 2017

Traffic lights contain very important information for safety driving. So, the delivery of the information to drivers in real-time is a very critical issue for advanced driver assistance systems. However, traffic light detection is quite difficult because of the small sized traffic lights and the occlusion in real world. In this paper, a traffic light detection method using modified color based saliency map and morphological information is proposed. It shows 98.14% of precisions and 83.52% of recalls on computer simulations.

• Journal of the Korea Computer Graphics Society
• /
• v.13 no.3
• /
• pp.1-6
• /
• 2007

Anticipation effect has been used as a traditional skill to enhance the dynamic motion of the traditional 2D animation. Basically, anticipation means the action of opposite direction which performs before the real action step. In this paper, we propose the perception-based video anticipation method to guide a user's visual attention to the important region. Using the image based attention map, we calculate the visual attention region and then combine this map with temporal saliency of video. We apply the anticipation effect in these saliency regions using the blur kernel. Using our method, we can generate the dynamic video motion which has attentive guidance.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.3
• /
• pp.298-303
• /
• 2014

In this paper, we proposed a robust object extraction and tracking algorithm in the IR image sequence acquired in the sea environment. In order to extract size-invariant object, we detect horizontal and vertical edges by using DWT and combine it to generate saliency map. To extract object region, binarization technique is applied to saliency map. The correspondences between objects in consecutive frames are defined by the calculating minimum weighted Euclidean distance as a matching measure. Finally, object trajectories are determined by considering false correspondences such as entering object, vanishing objects and false object and so on. The proposed algorithm can find trajectories robustly, which has shown by experimental results.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.7
• /
• pp.3245-3271
• /
• 2016

To obtain well-dehazed images at the receiver while sustaining low bit rates in the transmission pipeline, this paper investigates the effects of image dehazing methods using dehazing contrast-enhancement filters on image compression for surveillance systems. At first, this paper proposes a novel image dehazing method by using a new method of calculating the transmission function—namely, the direct denoising method. Next, we deduce the dehazing effects of the direct denoising method and image dehazing method based on dark channel prior (DCP) on image compression in terms of ringing artifacts and blocking artifacts. It can be concluded that the direct denoising method performs better than the DCP method for decompressed (reconstructed) images. We also improve the direct denoising method to obtain more desirable dehazed images with higher contrast, using the saliency map as the guidance image to modify the transmission function. Finally, we adjust the parameters of dehazing contrast-enhancement filters to obtain a corresponding composite peak signal-to-noise ratio (CPSNR) and blind image quality assessment (BIQA) of the decompressed images. Experimental results show that different filters have different effects on image compression. Moreover, our proposed dehazing method can strike a balance between image dehazing and image compression.