DOCSLIB.ORG

Compression and File Formats Compressing Moving Images

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Compression and File Formats Compressing Moving Images Compression and File Formats 1 Compressing Moving Images Methods: • Motion JPEG, Cinepak, Indeo, MPEG • Known as CODECs – compression / decompression algorithms – hardware and software implementations – symmetrical / asymmetrical Moving versus still images: • Temporal as well as spatial compression • Data transfer rates are critical – How much data is needed to maintain 24 frames/sec? Chapman & Chapman, Chapter 10 2 1 Spatial Compression • Compress each frame (still image) independently - also known as ‘intra-frame’ compression • Animated GIFs – single file containing multiple images compressed using LZW compression – only suitable for simple, cartoon-like animations • Motion JPEG (MJPEG) – compresses each frame using JPEG – suitable for videos of real-world action – implemented in hardware for real-time capture and compression of video – no standard file format! (MJPEG-A emerging) 3 Temporal Compression Also, known as ‘inter-frame’ compression - relies on only small changes between frames: 4 2 Temporal Compression Also Basic idea is this: • Certain frames designated as key frames – e.g. every sixth frame (may be user-settable) • Key frames either are left uncompressed or only spatially compressed • Each frame between key frames is replaced by a difference frame – records only differences between frame and preceding frame or preceding key frame • Difference frames readily compressed with RLE or dictionary-based schemes 5 Software CODECs for Multimedia • Multimedia is likely delivered on CD-ROM or over the Internet • Low to very low data transfer rates • High compression required • Software implementations of a variety of codecs – Cinepak – Intel Indeo – MPEG-1 6 3 Cinepak and Indeo • Similar codec methods that use vector quantization • Each frame divided into small rectangular blocks of pixels (the vectors) • Use a code book consisting of typical vectors (representing typical patterns that might occur in a image, e.g.) – areas of flat colour – sharp or soft edges – different textures – (the dictionary of still image compression) • Vector Quantization is the process of matching each image vector with its closest vector in the code book • Each image vector is replaced by its index in the code book • Image can be similarly re-constructed from these indexes and the codebook - de-compression efficient 7 Cinepak and Indeo (2) • Highly asymmetric: compression takes much longer than decompression – 150:1 for Cinepak • Also use temporal compression based on key and difference frames • full-motion full screen playback cannot be achieved with these codecs • Good enough for 320x240 pixels at 12 frames/sec – not great viewing! • New Sorenson codec does a bit better (can compress video to give a frame-rate of 50 kbytes/s) 8 4 MPEG-1 • Several international standards for compressed digital video are grouped together under the name, MPEG • Motion Picture Experts Group • Defines a data stream syntax and a decompressor • Compression is left to the software developer – “competitive advantage in the marketplace” • MPEG-1 Standard does define compression implicitly • So we can describe MPEG-1 compression which combines: • Spatial compression similar to JPEG, and • Temporal compression via difference frames and motion compensation. 9 Temporal Compression in MPEG-1 • Naïve approach subtracts each pixel in current frame from corresponding pixel in previous frame – result is 0 for pixels that do not change between frames – difference frames will contain large numbers of zeros and so will compress well, using e.g. RLE • BUT can possibly do better when there is large movement between frames – pictures usually composed of whole objects (e.g. a walking person) that move between frames • Difference frame would be better if we could match pixels from the moving object between frames • Motion compensation (if we can identify coherent area of change corresponding to object, then only need to record object’s movement together with changed pixels in the area) 10 5 Motion Compensation • Motion compensation in MPEG does not attempt to identify objects in a scene – hard computer vision / artificial intelligence problem!! • Divides each frame into 16 by 16 pixel blocks, known as macroblocks – different from the smaller blocks used in spatial compression • Attempts to predict where the corresponding macroblock is in the next frame – simple searching in a limited area around the macroblock (all possible displacements within a limited range are tried and best best match chosen) • Difference frame is then constructed by subtracting matching macroblocks (i.e. each macroblock from its predicted counterpart in next frame) • BUT also need to store displacement vectors (describing predicted displacement of macroblocks between frames)! 11 Frame Types in MPEG • Temporal compression has to start somewhere (e.g. Keyframe) • Key frames called I-pictures (I for intra) • Difference frames based on previous frames called P- pictures (predictive pictures) • MPEG also allows difference frames (motion compensation) based on later frames: B-pictures (bi-directional prediction) • A video clip can be encoded in compressed form as a sequence of I-, P- and B- pictures • MPEG Display order: IBBPBBIBBPBBI • Encoders normally use a repeating sequence, called Group of Pictures (GOP): IBBPBB (always starts with Key frame) • Display order is converted into a Bitstream order (for transmission): e.g. IPBBIBBPBBIBB – Decoder needs to reconstruct display order from bitstream 12 6 MPEG Variations MPEG-1: Delivers 1.2Mbps of video and 250Kbps of 2- channel audio from CD-ROM – best for multimedia applications – However, MPEG-1 cannot handle interlacing or HDTV formats, hence need for MPEG-2 for broadcasting & studio work MPEG-2: Needs higher data rates of 3-15Mbps, but delivers better picture quality, interlaced video formats and multichannel audio – designed for broadcast digital video (e.g. DVD) MPEG-4: emerging standard offering data rates from 5Kbps to 4Mbytesps, with interactive multimedia as well as video 13 New and Future CODECs • The challenge of sending video over the Internet has led to much research into new compression methods – very low bitrates required (e.g. 28kbps over a phoneline) • More sophisticated motion estimation • Wavelets, fractals etc • H.261 and H.263 – designed for video conferencing – use multiple compression techniques – “better than nothing” performance over a phoneline 14 7 File Formats for Video • Like file formats for still images, file formats for video provide a packaging that includes details of the contents of the file, such as the codec used • Various formats, such as... • Microsoft’s AVI • QuickTime • MPEG – not really a file format, just defines a bit (data) stream (like JPEG) 15 Video for Windows • AVI: Audio and Video Interleaved • Video and synchronized audio in a small window at around 15 frames-per-second without specialized hardware • Video is interleaved with the audio • Variety of codecs (identified by 4-byte code, FOURCC): – Cinepak (CVID) – Intel Indeo – Microsoft Video – Microsoft RLE (MRLE) – others 16 8 QuickTime • De facto standard (as of version 3.0 in 1998) • QuickTime (QT) movie (“.mov”) is a framework for organizing and manipulating the data that make up the actual video frames – many video codecs supported – can contain media other than video – media not necessarily stored in the .mov file • A movie plays according to a time base (which records rate at which it should be played back & its current position) – frames may be dropped to keep in time – enables constant playback speed on any system – maintains synchronization between video and audio 17 QuickTime • Success of QT is due to its Component-based architecture • Plug in new components (small programs conforming to certain interface conventions) to deal with new formats or provide new operations • Components include: – Compressors (for implementing e.g. Cinepak, Indeo) – Sequence grabbers: digitize video – Movie player component is used to provide a user-interface for playing movies: will handle AVI and MPEG – Transcoders: translate between data formats that use the same compression algorithm 18 9 Streamed Video • Playback as it arrives over the Internet • Current method of delivery on WWW: Progressive download or HTTP streaming – download movie to hard disk – start playing back when playback time of downloaded content equals download time of remainder of movie – still a large delay before playback begins and not suitable for “live” video • True streaming – never stored on user’s disk – must arrive at playback rate (not possible on dial-up internet connections using V90 modems or basic rate ISDN) – different movie formats depending on connection speed – Streaming QuickTime and RealVideo 19 File Formats for Animation • Macromedia’s Shockwave Flash – Native file format for Flash movies (and Director) – Vector images and animations – Metafile: can incorporate bitmap images and sound – De facto standard for simple animation on the Web – Browser plugins freely available • VRML – Virtual Reality Modelling Language – 3D interactive worlds – Browser plugins freely available 20 10 Shockwave Flash (SWF) • File contains two broad classes of items… • Definitions: dictionary of symbols that appear in the animation • Control items: instructions to place, remove or move a symbol – movement specified using transformation matrices • Rather like a program (cf. FIG, POVRay, VRML) – stored in compressed binary format: SMALL! • Flash creates simple animations • Director creates more complicated movies 21 VRML • Code directly, or output from 3D modelling package • Interactive 3D worlds: 22 11 VRML Example #VRML V1.0 ascii Separator { DirectionalLight { direction 0 0 -1 } PerspectiveCamera { position -8.6 2.1 5.6 orientation -0.1352 -0.9831 -0.1233 1.1417 focalDistance 10.84 } Separator { # The red sphere Material { diffuseColor 1 0 0 # Red Separator { # The blue cube } Material { Translation { translation 3 0 1 } diffuseColor 0 0 1 # Blue Sphere { radius 2.3 } } } Transform { translation -2.4 .2 1 rotation 0 1 1 .9 } Cube {} } 23 End of Lecture Exam revision/hints for my part of the course during tutorials next week! Good luck with all your exams! 24 12.
Load more

Recommended publications
  • Music 422 Project Report
    Music 422 Project Report Jatin Chowdhury, Arda Sahiner, and Abhipray Sahoo October 10, 2020 1 Introduction We present an audio coder that seeks to improve upon traditional coding methods by implementing block switching, spectral band replication, and gain-shape quantization. Block switching improves on coding of transient sounds. Spectral band replication exploits low sensitivity of human hearing towards high frequencies. It fills in any missing coded high frequency content with information from the low frequencies. We chose to do gain-shape quantization in order to explicitly code the energy of each band, preserving the perceptually important spectral envelope more accurately. 2 Implementation 2.1 Block Switching Block switching, as introduced by Edler [1], allows the audio coder to adaptively change the size of the block being coded. Modern audio coders typically encode frequency lines obtained using the Modified Discrete Cosine Transform (MDCT). Using longer blocks for the MDCT allows forbet- ter spectral resolution, while shorter blocks have better time resolution, due to the time-frequency trade-off known as the Fourier uncertainty principle [2]. Due to their poor time resolution, using long MDCT blocks for an audio coder can result in “pre-echo” when the coder encodes transient sounds. Block switching allows the coder to use long MDCT blocks for normal signals being encoded and use shorter blocks for transient parts of the signal. In our coder, we implement the block switching algorithm introduced in [1]. This algorithm consists of four types of block windows: long, short, start, and stop. Long and short block windows are simple sine windows, as defined in [3]: (( ) ) 1 π w[n] = sin n + (1) 2 N where N is the length of the block.
    [Show full text]
  • (A/V Codecs) REDCODE RAW (.R3D) ARRIRAW
    What is a Codec? Codec is a portmanteau of either "Compressor-Decompressor" or "Coder-Decoder," which describes a device or program capable of performing transformations on a data stream or signal. Codecs encode a stream or signal for transmission, storage or encryption and decode it for viewing or editing. Codecs are often used in videoconferencing and streaming media solutions. A video codec converts analog video signals from a video camera into digital signals for transmission. It then converts the digital signals back to analog for display. An audio codec converts analog audio signals from a microphone into digital signals for transmission. It then converts the digital signals back to analog for playing. The raw encoded form of audio and video data is often called essence, to distinguish it from the metadata information that together make up the information content of the stream and any "wrapper" data that is then added to aid access to or improve the robustness of the stream. Most codecs are lossy, in order to get a reasonably small file size. There are lossless codecs as well, but for most purposes the almost imperceptible increase in quality is not worth the considerable increase in data size. The main exception is if the data will undergo more processing in the future, in which case the repeated lossy encoding would damage the eventual quality too much. Many multimedia data streams need to contain both audio and video data, and often some form of metadata that permits synchronization of the audio and video. Each of these three streams may be handled by different programs, processes, or hardware; but for the multimedia data stream to be useful in stored or transmitted form, they must be encapsulated together in a container format.
    [Show full text]
  • Center for Signal Processing
    21 3. LOSSLESS/LOSSY COMPRESSION Overview of Lossless Compression: It is also known as: • Noiseless coding • Lossless coding • Invertible coding • Entropy coding • Data compaction codes. They can perfectly recover original data (if no storage or transmission bit errors, i.e., noiseless channel). 1. They normally have variable length binary codewords to produce variable numbers of bits per symbol. 2. Only works for digital sources. Properties of Variable rate (length) systems: • They can more efficiently trade off quality/distortion (noise) and rate. • They are generally more complex and costly to build. • They require buffering to match fixed-rate transmission system requirements. • They tend to suffer catastrophic transmission error propagation; in other words, once in error then the error grows quickly. 3. But they can provide produce superior rate/distortion (noise) trade-off. For instance, in image compression we can use more bits for edges, fewer for flat areas. Similarly, more bits can be assigned for plosives, fewer for vowels in speech compression. General idea: Code highly probable symbols into short binary sequences, low probability symbols into long binary sequences, so that average is minimized. Most famous examples: • Morse code: Consider dots and dashes as binary levels “0” and “1”. Assign codeword length inversely proportional to letter relative frequencies. In other words, most frequent letter (message) is assigned the smallest number of bits, whereas the least likely message has the longest codeword, such as “e” and “”z” in English alphabet. Total number of bits per second would much less than if we have used fixed-rate ASCII representation for alphanumerical data. • Huffman code 1952, employed in UNIX compact utility and many standards, known to be optimal under specific constraints.
    [Show full text]
  • Coding and Compression
    Chapter 3 Multimedia Systems Technology: Co ding and Compression 3.1 Intro duction In multimedia system design, storage and transport of information play a sig- ni cant role. Multimedia information is inherently voluminous and therefore requires very high storage capacity and very high bandwidth transmission capacity.For instance, the storage for a video frame with 640 480 pixel resolution is 7.3728 million bits, if we assume that 24 bits are used to en- co de the luminance and chrominance comp onents of each pixel. Assuming a frame rate of 30 frames p er second, the entire 7.3728 million bits should b e transferred in 33.3 milliseconds, which is equivalent to a bandwidth of 221.184 million bits p er second. That is, the transp ort of such large number of bits of information and in such a short time requires high bandwidth. Thereare two approaches that arepossible - one to develop technologies to provide higher bandwidth of the order of Gigabits per second or more and the other to nd ways and means by which the number of bits to betransferred can be reduced, without compromising the information content. It amounts to saying that we need a transformation of a string of characters in some representation such as ASCI I into a new string e.g., of bits that contains the same information but whose length must b e as small as p ossible; i.e., data compression. Data compression is often referred to as co ding, whereas co ding is a general term encompassing any sp ecial representation of data that achieves a given goal.
    [Show full text]
  • Experiment 7 IMAGE COMPRESSION I Introduction
    Experiment 7 IMAGE COMPRESSION I Introduction A digital image obtained by sampling and quantizing a continuous tone picture requires an enormous storage. For instance, a 24 bit color image with 512x512 pixels will occupy 768 Kbyte storage on a disk, and a picture twice of this size will not fit in a single floppy disk. To transmit such an image over a 28.8 Kbps modem would take almost 4 minutes. The purpose for image compression is to reduce the amount of data required for representing sampled digital images and therefore reduce the cost for storage and transmission. Image compression plays a key role in many important applications, including image database, image communications, remote sensing (the use of satellite imagery for weather and other earth-resource applications), document and medical imaging, facsimile transmission (FAX), and the control of remotely piloted vehicles in military, space, and hazardous waste control applications. In short, an ever-expanding number of applications depend on the efficient manipulation, storage, and transmission of binary, gray-scale, or color images. An important development in image compression is the establishment of the JPEG standard for compression of color pictures. Using the JPEG method, a 24 bit/pixel color images can be reduced to between 1 to 2 bits/pixel, without obvious visual artifacts. Such reduction makes it possible to store and transmit digital imagery with reasonable cost. It also makes it possible to download a color photograph almost in an instant, making electronic publishing/advertising on the Web a reality. Prior to this event, G3 and G4 standards have been developed for compression of facsimile documents, reducing the time for transmitting one page of text from about 6 minutes to 1 minute.
    [Show full text]
  • Ts 103 491 V1.1.1 (2017-04)
    ETSI TS 103 491 V1.1.1 (2017-04) TECHNICAL SPECIFICATION DTS-UHD Audio Format; Delivery of Channels, Objects and Ambisonic Sound Fields 2 ETSI TS 103 491 V1.1.1 (2017-04) Reference DTS/JTC-DTS-UHD Keywords audio, codec, object audio ETSI 650 Route des Lucioles F-06921 Sophia Antipolis Cedex - FRANCE Tel.: +33 4 92 94 42 00 Fax: +33 4 93 65 47 16 Siret N° 348 623 562 00017 - NAF 742 C Association à but non lucratif enregistrée à la Sous-Préfecture de Grasse (06) N° 7803/88 Important notice The present document can be downloaded from: http://www.etsi.org/standards-search The present document may be made available in electronic versions and/or in print. The content of any electronic and/or print versions of the present document shall not be modified without the prior written authorization of ETSI. In case of any existing or perceived difference in contents between such versions and/or in print, the only prevailing document is the print of the Portable Document Format (PDF) version kept on a specific network drive within ETSI Secretariat. Users of the present document should be aware that the document may be subject to revision or change of status. Information on the current status of this and other ETSI documents is available at https://portal.etsi.org/TB/ETSIDeliverableStatus.aspx If you find errors in the present document, please send your comment to one of the following services: https://portal.etsi.org/People/CommiteeSupportStaff.aspx Copyright Notification No part may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm except as authorized by written permission of ETSI.
    [Show full text]
  • (LSF) Vector Quantization in Speech Codec
    electronics Article An Efficient Codebook Search Algorithm for Line Spectrum Frequency (LSF) Vector Quantization in Speech Codec Yuqun Xue 1,2, Yongsen Wang 1,2, Jianhua Jiang 1,2, Zenghui Yu 1, Yi Zhan 1, Xiaohua Fan 1 and Shushan Qiao 1,2,* 1 Smart Sensing R&D Center, Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China; [email protected] (Y.X.); [email protected] (Y.W.); [email protected] (J.J.); [email protected] (Z.Y.); [email protected] (Y.Z.); [email protected] (X.F.) 2 Department of Microelectronics, University of Chinese Academy of Sciences, Beijing 100049, China * Correspondence: [email protected] Abstract: A high-performance vector quantization (VQ) codebook search algorithm is proposed in this paper. VQ is an important data compression technique that has been widely applied to speech, image, and video compression. However, the process of the codebook search demands a high computational load. To solve this issue, a novel algorithm that consists of training and encoding procedures is proposed. In the training procedure, a training speech dataset was used to build the squared-error distortion look-up table for each subspace. In the encoding procedure, firstly, an input vector was quickly assigned to a search subspace. Secondly, the candidate code word group was obtained by employing the triangular inequality elimination (TIE) equation. Finally, a partial distortion elimination technique was employed to reduce the number of multiplications. The proposed method reduced the number of searches and computation load significantly, especially when the input vectors were uncorrelated.
    [Show full text]
  • Multi-Image Classification and Compression Using Vector Quantization
    W&M ScholarWorks Dissertations, Theses, and Masters Projects Theses, Dissertations, & Master Projects 1999 Multi-image classification and compression using vector quantization Beverly J. Thompson College of William & Mary - Arts & Sciences Follow this and additional works at: https://scholarworks.wm.edu/etd Part of the Computer Sciences Commons Recommended Citation Thompson, Beverly J., "Multi-image classification and compression using vector quantization" (1999). Dissertations, Theses, and Masters Projects. Paper 1539623956. https://dx.doi.org/doi:10.21220/s2-qz52-5502 This Dissertation is brought to you for free and open access by the Theses, Dissertations, & Master Projects at W&M ScholarWorks. It has been accepted for inclusion in Dissertations, Theses, and Masters Projects by an authorized administrator of W&M ScholarWorks. For more information, please contact [email protected]. INFORMATION TO USERS This manuscript has baen reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter fece, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these win be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps.
    [Show full text]
  • Algorithms for Fast Vector Quantization∗
    Algorithms for Fast Vector Quantization∗ Sunil Aryay Department of Computer Science The Hong Kong University of Science and Technology Clear Water Bay, Kowloon, Hong Kong David M. Mountz Department of Computer Science and Institute for Advanced Computer Studies University of Maryland, College Park, Maryland, USA Abstract Nearest neighbor searching is an important geometric subproblem in vector quanti- zation. Existing studies have shown that the difficulty of solving this problem efficiently grows rapidly with dimension. Indeed, existing approaches on unstructured codebooks in dimension 16 are little better than brute-force search. We show that if one is willing to relax the requirement of finding the true nearest neighbor then dramatic improve- ments in running time are possible, with negligible degradation in the quality of the result. We present an empirical study of three nearest neighbor algorithms on a number of data distributions, and in dimensions varying from 8 to 16. The first algorithm is the standard k-d tree algorithm which has been enhanced to use incremental distance calculation, the second is a further improvement that orders search by the proximity of the k-d cell to the query point, and the third is based on a simple greedy search in a structure called a neighborhood graph. Key words: Nearest neighbor searching, closest-point queries, data compression, vec- tor quantization, k-d trees. 1 Introduction The nearest neighbor problem is to find the point closest to a query point among a set of n points in d-dimensional space. We assume that the distances are measured in the Euclidean metric. Finding the nearest neighbor is a problem of significant importance in many applications.
    [Show full text]
  • Vq-Wav2vec:Self-Supervised Learningof Discrete Speech Representations
    Published as a conference paper at ICLR 2020 VQ-WAV2VEC:SELF-SUPERVISED LEARNING OF DISCRETE SPEECH REPRESENTATIONS Alexei Baevski∗4 Steffen Schneider∗5y Michael Auli4 4 Facebook AI Research, Menlo Park, CA, USA 5 University of Tubingen,¨ Germany ABSTRACT We propose vq-wav2vec to learn discrete representations of audio segments through a wav2vec-style self-supervised context prediction task. The algorithm uses either a Gumbel-Softmax or online k-means clustering to quantize the dense representations. Discretization enables the direct application of algorithms from the NLP community which require discrete inputs. Experiments show that BERT pre-training achieves a new state of the art on TIMIT phoneme classification and WSJ speech recognition.1 1 INTRODUCTION Learning discrete representations of speech has gathered much recent interest (Versteegh et al., 2016; Dunbar et al., 2019). A popular approach to discover discrete units is via autoencoding (Tjandra et al., 2019; Eloff et al., 2019; Chorowski et al., 2019) sometimes coupled with an autoregressive model (Chung et al., 2019). Another line of research is to learn continuous speech representations in a self-supervised way via predicting context information (Chung & Glass, 2018; van den Oord et al., 2018; Schneider et al., 2019). In this paper, we combine these two lines of research by learning discrete representations of speech via a context prediction task instead of reconstructing the input. This enables us to directly apply well performing NLP algorithms to speech data (Figure 1a). L1 L2 L3 vq-wav2vec BERT AM t C h e Z^ q c Z a X t (a) vq-wav2vec (b) Discretized speech training pipeline Figure 1: (a) The vq-wav2vec encoder maps raw audio (X ) to a dense representation (Z) which is quantized (q) to Z^ and aggregated into context representations (C); training requires future time step prediction.
    [Show full text]
  • Video Compression: MPEG-4 and Beyond Video Compression: MPEG-4 and Beyond
    Video Compression: MPEG-4 and Beyond Video Compression: MPEG-4 and Beyond Ali Saman Tosun, [email protected] Abstract Technological developments in the networking technology and the computers make use of video possible. Storing and transmitting uncompressed raw video is not a good idea, it requires large storage space and bandwidth. Special algorithms which take the characteristics of the video into account can compress the video with high compression ratio. In this paper I will give a overview of the standardization efforts on video compression: MPEG-1, MPEG-2, MPEG-4, MPEG-7, and I will explain the current video compression trends briefly. See also: Multimedia Networking Products | Multimedia Over IP: RSVP, RTP, RTCP, RTSP | Multimedia Networking References | Books on Multimedia | Protocols for Multimedia on the Internet (Class Lecture) | Video over ATM networks | Multimedia networks: An introduction (Talk by Prof. Jain) | Other Reports on Recent Advances in Networking Back to Raj Jain's Home Page Table of Contents: ● 1. Introduction ● 2. H.261 ● 3. H.263 ● 4. H.263+ ● 5. MPEG ❍ 5.1 MPEG-1 ❍ 5.2 MPEG-2 ❍ 5.3 MPEG-3 ❍ 5.4 MPEG-4 ❍ 5.5 MPEG-7 ● 6. J.81 ● 7. Fractal-Based Coding ● 8. Model-based Video Coding http://www.cis.ohio-state.edu/~jain/cis788-99/compression/index.html (1 of 13) [2/7/2000 10:39:09 AM] Video Compression: MPEG-4 and Beyond ● 9. Scalable Video Coding ● 10 . Wavelet-based Coding ● Summary ● References ● List of Acronyms 1. Introduction Over the last couple of years there has been a great increase in the use of video in digital form due to the popularity of the Internet.
    [Show full text]
  • A Novel and Efficient Vector Quantization Based CPRI
    1 A Novel and Efficient Vector Quantization Based CPRI Compression Algorithm Hongbo Si, Boon Loong Ng, Md. Saifur Rahman, and Jianzhong (Charlie) Zhang Abstract The future wireless network, such as Centralized Radio Access Network (C-RAN), will need to deliver data rate about 100 to 1000 times the current 4G technology. For C-RAN based network architecture, there is a pressing need for tremendous enhancement of the effective data rate of the Common Public Radio Interface (CPRI). Compression of CPRI data is one of the potential enhancements. In this paper, we introduce a vector quantization based compression algorithm for CPRI links, utilizing Lloyd algorithm. Methods to vectorize the I/Q samples and enhanced initialization of Lloyd algorithm for codebook training are investigated for improved performance. Multi-stage vector quantization and unequally protected multi-group quantization are considered to reduce codebook search complexity and codebook size. Simulation results show that our solution can achieve compression of 4 times for uplink and 4:5 times for downlink, within 2% Error Vector Magnitude (EVM) distortion. Remarkably, vector quantization codebook proves to be quite robust against data modulation mismatch, fading, signal-to- noise ratio (SNR) and Doppler spread. I. INTRODUCTION arXiv:1510.04940v1 [cs.IT] 16 Oct 2015 The amount of wireless IP data traffic is projected to grow by well over 100 times within a decade (from under 3 exabytes in 2010 to more than 500 exabytes by 2020) [1]. To address such wireless data traffic demand, there has been increasing effort to define the 5G network in recent years. It is widely recognized that the 5G network will be required to deliver data rate about The material in this paper was submitted in part to the 2015 IEEE Global Communications Conference (Globecom 2015), San Diego, CA, USA, Dec.
    [Show full text]