Search

With proliferation of social media, image has become ubiquitous giving rise to the demand and importance of image semantic analysis and retrieval to access information quickly on social media. However, even with humongous information available, there are certain categories of images which are important for certain applications but are very scarce. Convolutional neural network is an effective method to extract high-level semantic features for image database retrieval. To overcome the problem of over-fitting when the number of training samples in dataset is limited, this paper proposes an image database retrieval algorithm based on the framework of transfer learning and feature fusion. Based on the fine-tuning of the pre-trained Convolutional Neural Network (CNN), the proposed algorithm first extracts the semantic features of the images. Principal Component Analysis (PCA) is then applied for dimension reduction and to reduce the computational complexity. Last, the semantic feature extracted from the CNN is fused with traditional low-level visual feature to improve the retrieval accuracy further. Experimental results demonstrated the effectiveness of the proposed method for image database retrieval. [ABSTRACT FROM AUTHOR]

We propose a saturation-aware human attention region-of-interest (SA-HAROI) video compression method that performs a perceptual adaptive quantization algorithm on video frames as a function of the distribution of their luminance, motion vector, and color saturation. Our work is an application of a psycho-visual study that demonstrated that human attention automatically enhanced perceived saturation. Consequently, the adaptive quantization phase of our compression algorithm is characterized by a luminance and saturation-aware just noticeable distortion (JND) function. After running multiple experiments on 18 videos with various resolutions ranging from QCIF to 4 K, results showed that our method achieves higher compression than that of both the H.264/AVC JM and the HEVC HM while maintaining subjective quality. We observed that in comparison to both implementation of the standards (JM and HM), for an IPPP coding structure, the performance of our algorithm culminated with HD and 4 K videos yielding a bit rate reduction averaging 15% and an encoding time reduction of about 20% in certain cases. Finally, after comparing our method to other similar techniques, we concluded that saturation is a significant parameter in the improvement of video compression. [ABSTRACT FROM AUTHOR]

Traditional visual saliency based video compression methods try to encode the image with higher quality in the region of saliency. However, the saliency feature changes according to persons, viewpoints, and distances. In this paper, we propose to apply a technique of human centered perceptual computation to improve video coding in the region of human centered perception. To detect the region of interest (ROI) of human body, upper body, frontal face, and profile face, we construct Harr and histogram of oriented gradients features based combo of detectors to analyze a video in the first frame (intra-frame). From the second frame (inter-frame) onward, the optical flow image is computed in the ROI area of the first frame. The optical flow in human centered ROI is then used for macroblock (MB) quantization adjustment in H.264/AVC. For each MB, the quantization parameter (QP) is optimized with density value of optical flow image. The QP optimization process is based on a MB mapping model, which can be calculated by an inverse of the inverse tangent function. The Lagrange multiplier in the rate distortion optimization is also adapted so that the MB distortion at human centered region is minimized. We apply our technique to the H.264 video encoder to improve coding visual quality. By evaluating our scheme with the H.264 reference software, our results show that the proposed algorithm can improve the visual quality of ROI by about 1.01 dB while preserving coding efficiency. [ABSTRACT FROM AUTHOR]

VC-1 is one of the three video coding standards for Blu-ray DVD, which also includes MPEG-2 and H.264. In this paper, an efficient transcoding algorithm from VC-1 video to H.264 video is discussed, we present a transcoder that addresses I, P, B, and interlaced pictures. The main differences between the two standards are analyzed and a solution for transforming from one to the other is presented. The paper proposes a pixel domain transcoder which exploits the variable size transform used in VC-1 to select the variable block size for motion compensation in H.264. We also discuss transcoding of high profile video features; in particular, adaptive size transform and interlacing. Experimental results show that the pixel domain approach reduces computational complexity by more than 50 % as compared to a cascaded one with negligible drop in peak-signal-to-noise-ratio (PSNR). [ABSTRACT FROM AUTHOR]

An introduction is presented in which the editor discusses various reports within the issue on topics including distributed systems for multimedia content production; and a collusion-secure forensic mark design technique for protecting copyrights on digital content; and motion sensing technologies.

In transcoding, quantization and other techniques could result in lower video output quality. To address this problem a novel super-resolution (SR) algorithm based on irregular sampling (IS) is presented in this paper. The high-resolution (HR) frame is obtained as an interpolation of one or more previous frames; the resulting interpolated frame has samples non-uniformly spaced in the areas where movement happened. To reconstruct the irregular sampled frame we use a well-known irregular sampling algorithm modified to perform in 2-D space. Moreover, because SR algorithms are in general computationally expensive, we also present a hardware feasibility study. The proposed solution does not target any specific application but we have specifically tested the algorithm in a transcoding environment. In particular, we have applied it to VC-1 to H.264 transcoding and applied down/up sampling. Experimental results show that the proposed algorithm improves video quality significantly. [ABSTRACT FROM AUTHOR]

Conventional statistical multiplexing methods for MPEG-1/2 programs involve high computationally complex transcoding (decoding and re-encoding) process to convert the original bit-rate into the target bit-rate. To avoid a time-consuming transcoding process, a new statistical multiplexer is proposed to enable content provider to easily convert the bit-rates by exploiting the MPEG-4 fine granularity scalability (FGS) coding scheme. The proposed statistical multiplexer is particularly useful for multiple-program broadcasting applications, including Internet television and video on demand, as well as value-added MPEG-4 video streaming services for DVB and ATSC digital TV systems. The proposed multiplexer mainly includes two parts: the FGS-based frame lag scheme and the optimal bit-plane truncation scheme. The FGS-based frame lag scheme exploits intra- and inter-layer correlations exist in MPEG-4 FGS bit-streams. The optimal bit-plane truncation scheme dynamically truncates enhancement layers of FGS bit-streams under the available bandwidth constraint and the quality/smoothness constraint. Experimental results show that high statistical multiplexing efficiency, inter-program fairness, and intra-program smoothness are achieved by the proposed multiplexer. [ABSTRACT FROM AUTHOR]

In this paper, an architecture for real-time digital HDTV video decoding is presented. Our architecture is based on a dual decoding datapath controlled in a fixed schedule with an efficient write-back scheme for anchor pictures. The decoding datapath is synchronized at the block (8 × 8 pixels) level. Unlike other decoding approaches such as the slice bar decoding method and the cross-divide method, our scheme reduces memory access contention problem to achieve real-time HDTV decoding without a high cost in overall decoder buffers, architecture, and bus. In comparison to data-flow approaches, our method eliminates the complexity associated with tagged data operations. Our anchor picture storage is organized to minimize page-breaks during memory accesses. Simulation shows that with a relatively low rate 81 MHz clock, our decoder can decode MPEG-2 MP@HL HDTV in real-time, based on an ATSC video format of 1,920 × 1,080 pixels/frame at 30 frames/s, at a bit rate of 18 to 20 Mbps. [ABSTRACT FROM AUTHOR]

In verifying the design correctness of a specific class of architecture, special purpose formal design verifier has the advantage of being able to exploit the attributes of that architecture class to produce efficiency in the design verification process. Such development is important due to the fact that architecture design verification using general purpose theorem prover is usually extremely time consuming. This paper briefly presents the techniques and a Prolog-based verifier VSTA that we developed for formal design verification of systolic array architectures in DSP applications. Systolic architecture has been a popular class of parallel architecture due to its suitability for VLSI implementation. The techniques developed are discussed and a formalism (STA) developed for specifying and verifying systolic designs is reviewed. The strategies and notation adopted exploit systolic attributes for fast design verifications. We develop a Prolog-based verifier to automate our techniques due to Prolog's powerful pattern matching and automatic back-tracking mechanism, its popularity and quality, its similarity in representing facts with STA, and its wide acceptance for lower level module and circuit verification (so as to achieve multilevel reasoning later). In the paper, we also describe the application for our tool to verify the correctness of two systolic array designs. Executing the verifier on our workstation shows that a typical array design can usually be verified in less than 10 minutes. [ABSTRACT FROM AUTHOR]

The article discusses various reports published within the issue, including one on a fast hypothesis motion compensated filter for video denoising, another on the use of a super-resolution algorithm based on irregular sampling for video transcoding with resolution conversion, as well as a report on pipelining multi-channel central caching, a novel network-on-chip architecture, to address the cost and communication latency/throughput of existing architectures.