| Title | People Detection Using Feature Vector Matching |
| Author | *Hwa-Young Kim, Rae-Hong Park (Sogang University, Republic of Korea) |
| Page | pp. 449 - 452 |
| Keyword | edge orientation map, crowd segmentation, people detection, surveillance |
| Abstract | This paper presents a novel method to detect people in real time from a video sequence taken from a single fixed camera. Foreground images are obtained from a sequence of operations: background differencing, thresholding, and morphological operation. We make use of quantized edge orientation to represent shape of human because human has particular edge orientation distribution on each of body parts. Then, a feature vector derived from the edge orientation map is computed in each of body parts of foreground images. The dataset, composed of sets of feature vectors, is used for matching with feature vectors computed from the current image of the input sequence. We use the K-nearest neighbors to match the feature vectors. The proposed people detection method uses a proper feature vector that represents human edge orientation distribution, and employs simple matching steps for real-time processing. Experimental results with a number of test sequences demonstrate the effectiveness of the proposed method. |
| Title | Bilateral Motion Estimation Based Frame Error Concealment for Video Transmission over Wireless Networks |
| Author | Dinh Trieu Duong, Jun-Hyung Kim, Sung-Hyun Cheon, *Yeo-Jin Yoon, Sung-Jea Ko (Korea University, Republic of Korea) |
| Page | pp. 453 - 456 |
| Keyword | Error concealment, motion estimation |
| Abstract | In this paper, we propose a new error concealment algorithm that
can effectively reconstruct the lost frame caused by transmission errors
in the wireless environment. The proposed algorithm employs the bilateral motion estimation scheme where the received motion vectors (MVs) in the neighboring frames are weighted and utilized to construct the MV field for the concealed frame. Experimental results show that the proposed algorithm outperforms other techniques in term of both peak signal-to-noise ratio (PSNR) performance and subjective visual quality. |
| Title | 8x8 transformation based all zero block detection for H.264/AVC encoder |
| Author | *Jia Su, Qin Liu, Satoshi Goto, Takeshi Ikenaga (Graduate School of Information, Production and Systems, Waseda University, Japan) |
| Page | pp. 461 - 464 |
| Keyword | 8x8 integer transform and quantization, all zero block, H.264/AVC, sum of absolute difference (SAD) |
| Abstract | By introducing all zero block (AZB) algorithm, this paper provides an easily implement solution for reducing the complexity of an 8x8 integer transform and quantization. Through the statistic analysis of all zero block for more than 30 typical sequences, We set cubic modules for threshold decision of QCIF, CIF and 720p which can almost reach 100% accuracy and averagely save 5%~40% (according to different sequences and QP values) of the 8x8 DCT and quantization calculation.
Index Terms—8x8 integer transform and quantization, all zero block, H.264/AVC, sum of absolute difference (SAD). |
| Title | Multi-camera Imaging System Using Super-Resolution |
| Author | Jong Hyun Park, *Hyun Mook Oh, Moon Gi Kang (Yonsei University, Republic of Korea) |
| Page | pp. 465 - 468 |
| Keyword | Imaging and Image Processing, Digital Imaging, Super-Resolution, Image Restoration, Interpolation |
| Abstract | This work presents a high-resolution image acquisition system by
combining a multi-camera imaging system composed of multiple
low-resolution digital cameras and a color super-resolution image
reconstruction algorithm. Aligned and adjusted multiple mini-cameras
and the controlling system consist the multi-camera imaging system.
Some of the cameras obtain intensity information and the other ones
receive color information. The spatial intensity information is
enhanced by the Bayesian super-resolution image reconstruction
algorithm. The high-resolution intensity data and the color channel
inputs are merged to generate the final color composite
high-resolution image. With experimental results, the performance of
the proposed system is assessed. |