Jun-Ren Ding

Two-layer and adaptive entropy coding algorithms for H.264-based lossless image coding

In this paper, we propose two-layer coding algorithm to improve the performance of the H.264-based lossless (H.264-LS) image coding. From universal access point of view, the proposed method is based on the H.264 lossy image coding with other CABAC layer to compensate the lossy portion. Besides, the H.264-LS with DPCM (H264-LS_DPCM) and H.264-LS achieve different coding performance in use of CABAC and CAVLC entropy coders without the DCT and quantization. We further suggest an adaptive entropy coding (AEC) algorithm to determine the best entropy coder by using the image content variations, which is calculated from the sum of absolute difference of intra prediction. Simulation results show that the proposed AEC method have good correct detection rates and improvement of compression rate for H.264-LS and H.264-LS_DPCM coders. The two-layer H.264-LS almost have the same compression rate than the H.264-LS_DPCM.

A SIMPLIFIED SPIHT ALGORITHM

Abstract Based on the SPIHT algorithms, we define two modifications to develop a simpler image coding method. The first concept is obtained from the relationship between the bit‐planes and the target bit‐rate. The second concept is obtained from the relationship between the initial threshold and the target bit‐rate. Based on the abovementioned concepts, we can discard the refinement pass and improve the image quality at different target bit‐rates. The simulation results show that in comparison with the original SPIHT algorithm, the proposed algorithm can reduce memory usage by approximately 50% and computation time by approximately 30% with an acceptable PSNR loss.

Motion-based Adaptive GOP Algorithms for Efficient H.264/AVC Compression

H.264/AVC provides several advanced features such as improved coding efficiency and error robustness for video storage and transmission. In this paper, to further improve the coding quality for encoding possible scene changes and fast variation sequences, we propose a new adaptive GOP scheme that uses the existing motion vectors and motion residuals. The proposed scheme effectively identifies the H.264/AVC coding core {I, P} frames using a statistical table that improves the coding efficiency for changed scene sequences and different video content variance. Simulation results show that the proposed adaptive GOP scheme can increase PSNR 0.65 dB on average and achieve a 92% correct rate for scene change detection from the original H.264 reference software JM10.1 operated in the fixed GOP size.

Image-based auto-positioning brush for LCD displays

The interactive devices, which help humans to easily access the computer and video game player, become an important technology to provide interactive multimedia services and entertainments in the 21st century. Considering the cost and the effectiveness of the human machine interface (HMI) devices, in this paper, we design a low cost image-based auto-positioning brush (IA-Brush), which can automatically locate the position of the window and modify the affection of angle displacement of camera. The proposed IA-Brush based on a low cost video camera can be easily implemented for low cost liquid crystal display monitor by using image processing techniques figure with a designated background grids. Simulation results with exhibition of interactive painting games show that we can achieve a low cost and effective HMI by using a video camera only.

Adaptive entropy coding with (5, 3) DWT for H.264 lossless image coding

Lossless image coding is a important research field for multimedia store and digital library. H.264 provides a better compression rate than the JPEG-LS based on intra prediction and entropy coding (CABAC and CAVLC). In this paper, we suggest that entropy coding should be adaptively adopted for various image content variations(ICV) to increase H.264-LS compression rate. The ICV is obtained by existing intra prediction residue values in our proposed method. In the meantime, we add the (5, 3) discrete wavelet transform (DWT) for higher ICV images after executing 4times4 intra prediction to increase H.264-LS compression rate again. Simulation results show that using the adaptive entropy coding has 97% correct rate and 8.7% compression rate improvement than used single entropy coding. If add the (5, 3) DWT filtering again, we can at least obtain 2.3% compression rate than the adaptive entropy coding.

Fast adaptive GOP design for H.264/SVC

In order to concurrently provide different video services for diverse access devices, the scalable quality and complexity services in a single bit stream offered by H.264-based scalable video coding (H.264/SVC) has gained a lot of attentions recently. However, the motion-compensated temporal filtering (MCTF) structure consumes a large amount of operation time when combined with Core Experiment 2 (CE2), which implements adaptive GOP detection (AGD). In this paper, we suggest a video content variation (VCV) feature to implement fast AGD method by using block-based classification before executing MCTF. Based on the JSVM 7.0 software, the simulation results show that the proposed method reduces about 67% of the operation time consumed on average as compared to the CE2.

Joint Adaptive GOP and SCD Coding for Improving H.264 Scalable Video Coding

In order to concurrently provide different video services for diverse access devices, the scalable quality and complexity services in a single bit stream offered by H.264-based scalable video coding (H.264/SVC) has gained a lot of attentions recently. However, the motion-compensated temporal filtering (MCTF) structure consumes a large amount of operation time when combined with Core Experiment 2 (CE2), which implements adaptive GOP detection (AGD). Besides, a simple scene change detection (SCD) method is required to decrease the mis-position motion estimation and yield robust AGD. In this paper, we suggest a video content variation (VCV) feature to implement simple AGD and SCD methods by using block-based classification before executing MCTF. Based on the JSVM 7.0 software, the simulation results show that the proposed method not only reduces about 67% of the operation time consumed but also improves about 0.1 dB PSNR on average as compared to the CE2, where the proposed SCD can attain a detection rate of 89% to help for performance improvement.

Personal gesture-driven virtual walk-through systems

In this paper, we propose the use of the body gesture, instead of mouse and keyboard, as an interface to browse the exhibited multimedia video contents. For real applications, we have defined some control modes, which are based on human natural movements, to construct the commands of interactivities. These commands controlled a body gesture, are able to be built to control a virtual reality system for enhancing users experiences of exhibitions. In our demo system, this body gesture offers a more nature and intuitive interface for the users, who easily walk through a virtual environment, which is created by specific oil paintings generated by graph engines. The proposed system can help the viewers to obtain more information about those paintings such that we can enhance their exploration experiences in touring multimedia contents.