157:
101:. As can be seen from the images, the bottom (motion compensated) difference between two frames contains significantly less detail than the prior images, and thus compresses much better than the rest. Thus the information that is required to encode compensated frame will be much smaller than with the difference frame. This also means that it is also possible to encode the information using difference image at a cost of less compression efficiency but by saving coding complexity without motion compensated coding; as a matter of fact that motion compensated coding (together with
1288:
142:
127:
418:(OBMC) is a good solution to these problems because it not only increases prediction accuracy but also avoids blocking artifacts. When using OBMC, blocks are typically twice as big in each dimension and overlap quadrant-wise with all 8 neighbouring blocks. Thus, each pixel belongs to 4 blocks. In such a scheme, there are 4 predictions for each pixel which are summed up to a weighted mean. For this purpose, blocks are associated with a window function that has the property that the sum of 4 overlapped windows is equal to 1 everywhere.
27:
2580:
2570:
383:(VBSMC) is the use of BMC with the ability for the encoder to dynamically select the size of the blocks. When coding video, the use of larger blocks can reduce the number of bits needed to represent the motion vectors, while the use of smaller blocks can result in a smaller amount of prediction residual information to encode. Other areas of work have examined the use of variable-shape feature metrics, beyond block boundaries, from which interframe vectors can be calculated. Older designs such as
600:
and
Jaswant R. Jain further developed motion-compensated DCT video compression, also called block motion compensation. This led to Chen developing a practical video compression algorithm, called motion-compensated DCT or adaptive scene coding, in 1981. Motion-compensated DCT later became the standard
327:
The main disadvantage of block motion compensation is that it introduces discontinuities at the block borders (blocking artifacts). These artifacts appear in the form of sharp horizontal and vertical edges which are easily spotted by the human eye and produce false edges and ringing effects (large
361:
Frames can also be predicted from future frames. The future frames then need to be encoded before the predicted frames and thus, the encoding order does not necessarily match the real frame order. Such frames are usually predicted from two directions, i.e. from the I- or P-frames that immediately
421:
Studies of methods for reducing the complexity of OBMC have shown that the contribution to the window function is smallest for the diagonally-adjacent block. Reducing the weight for this contribution to zero and increasing the other weights by an equal amount leads to a substantial reduction in
46:
files. Motion compensation describes a picture in terms of the transformation of a reference picture to the current picture. The reference picture may be previous in time or even from the future. When images can be accurately synthesized from previously transmitted/stored images, the compression
434:
In motion compensation, quarter or half samples are actually interpolated sub-samples caused by fractional motion vectors. Based on the vectors and full-samples, the sub-samples can be calculated by using bicubic or bilinear 2-D filtering. See subclause 8.4.2.2 "Fractional sample interpolation
371:
Further, the use of triangular tiles has also been proposed for motion compensation. Under this scheme, the frame is tiled with triangles, and the next frame is generated by performing an affine transformation on these triangles. Only the affine transformations are recorded/transmitted. This is
308:
To exploit the redundancy between neighboring block vectors, (e.g. for a single moving object covered by multiple blocks) it is common to encode only the difference between the current and previous motion vector in the bit-stream. The result of this differentiating process is mathematically
263:
MPEG-4 ASP supports global motion compensation with three reference points, although some implementations can only make use of one. A single reference point only allows for translational motion which for its relatively large performance cost provides little advantage over block based motion
82:
of a movie, the only difference between one frame and another is the result of either the camera moving or an object in the frame moving. In reference to a video file, this means much of the information that represents one frame will be the same as the information used in the next frame.
572:
coding in the spatial dimension. In 1975, John A. Roese and Guner S. Robinson extended Habibi's hybrid coding algorithm to the temporal dimension, using transform coding in the spatial dimension and predictive coding in the temporal dimension, developing
324:, used by H.264 and MPEG-4/ASP) is used. The computational expense of sub-pixel precision is much higher due to the extra processing required for interpolation and on the encoder side, a much greater number of potential source blocks to be evaluated.
620:
standard was developed in 1988 based on motion-compensated DCT compression, and it was the first practical video coding standard. Since then, motion-compensated DCT compression has been adopted by all the major video coding standards (including the
464:
lead to bleeding between adjacent pixels. If no higher internal resolution is used the delta images mostly fight against the image smearing out. The delta image can also be encoded as wavelets, so that the borders of the adaptive blocks match.
422:
complexity without a large penalty in quality. In such a scheme, each pixel then belongs to 3 blocks rather than 4, and rather than using 8 neighboring blocks, only 4 are used for each block to be compensated. Such a scheme is found in the
37:
in computing is an algorithmic technique used to predict a frame in a video given the previous and/or future frames by accounting for motion of the camera and/or objects in the video. It is employed in the encoding of video data for
86:
Using motion compensation, a video stream will contain some full (reference) frames; then the only information stored for the frames in between would be the information needed to transform the previous frame into the next frame.
301:). Each block is predicted from a block of equal size in the reference frame. The blocks are not transformed in any way apart from being shifted to the position of the predicted block. This shift is represented by a
258:
A straight line (in the time direction) of pixels with equal spatial positions in the frame corresponds to a continuously moving point in the real scene. Other MC schemes introduce discontinuities in the time
30:
Visualization of MPEG block motion compensation. Blocks that moved from one frame to the next are shown as white arrows, making the motions of the different platforms and the character clearly visible.
358:). The source blocks typically overlap in the source frame. Some video compression algorithms assemble the current frame out of pieces of several different previously transmitted frames.
814:
Aizawa, Kiyoharu, and Thomas S. Huang. "Model-based image coding advanced video coding techniques for very low bit-rate applications." Proceedings of the IEEE 83.2 (1995): 259-271.
581:(FFT), developing inter-frame hybrid coders for both, and found that the DCT is the most efficient due to its reduced complexity, capable of compressing image data down to 0.25-
805:
Zeng, Kai, et al. "Characterizing perceptual artifacts in compressed video streams." IS&T/SPIE Electronic
Imaging. International Society for Optics and Photonics, 2014.
616:(now ITU-T) in 1984. H.120 used motion-compensated DPCM coding, which was inefficient for video coding, and H.120 was thus impractical due to low performance. The
297:(MC DCT), is the most widely used motion compensation technique. In BMC, the frames are partitioned in blocks of pixels (e.g. macro-blocks of 16Ă—16 pixels in
493:
A precursor to the concept of motion compensation dates back to 1929, when R.D. Kell in
Britain proposed the concept of transmitting only the portions of an
760:
Proceedings of the
International Computer Conference 2006 on Wavelet Active Media Technology and Information Processing: Chongqing, China, 29-31 August 2006
901:
788:
1079:
Roese, John A.; Robinson, Guner S. (30 October 1975). Tescher, Andrew G. (ed.). "Combined
Spatial And Temporal Coding Of Digital Image Sequences".
878:. Joint Video Team (JVT) of ISO/IEC MPEG & ITU-T VCEG (ISO/IEC JTC1/SC29/WG11 and ITU-T SG16 Q.6). July 2002. pp. 11, 24–9, 33, 40–1, 53–6
529:(MC DCT) coding, also called block motion compensation (BMC) or DCT motion compensation. This is a hybrid coding algorithm, which combines two key
252:
It models the dominant motion usually found in video sequences with just a few parameters. The share in bit-rate of these parameters is negligible.
1367:
204:
are more complex because the image sequence must be transmitted and stored out of order so that the future frame is available to generate the
1404:
597:
95:
The following is a simplistic illustrated explanation of how motion compensation works. Two successive frames were captured from the movie
70:
formats, typically use motion-compensated DCT hybrid coding, known as block motion compensation (BMC) or motion-compensated DCT (MC DCT).
2072:
1883:
1356:
313:
will take advantage of the resulting statistical distribution of the motion vectors around the zero vector to reduce the output size.
1772:
1223:
1153:
Chen, Wen-Hsiung; Smith, C. H.; Fralick, S. C. (September 1977). "A Fast
Computational Algorithm for the Discrete Cosine Transform".
2278:
2101:
1895:
937:
568:
introduced hybrid coding, which combines predictive coding with transform coding. However, his algorithm was initially limited to
1586:
2283:
1860:
460:
uses wavelets, and these can also be used to encode motion without gaps between blocks in an adaptive way. Fractional pixel
320:. The in-between pixels are generated by interpolating neighboring pixels. Commonly, half-pixel or quarter pixel precision (
2013:
1129:
210:
After predicting frames using motion compensation, the coder finds the residual, which is then compressed and transmitted.
1155:
942:
2629:
2614:
2390:
2128:
2067:
1878:
1828:
1651:
329:
1511:
1496:
1397:
1331:
1309:
1231:
1193:
1137:
1060:
772:
732:
565:
1302:
2503:
454:
is often considered as the third dimension. Still, image coding techniques can be expanded to an extra dimension.
2513:
2351:
2202:
2121:
1915:
658:
363:
196:
188:
2486:
2106:
1900:
1688:
2583:
1619:
869:
2248:
354:
is divided up into non-overlapping blocks, and the motion compensation vectors tell where those blocks move
2573:
2476:
2018:
1576:
1390:
911:
577:
motion-compensated hybrid coding. For the spatial transform coding, they experimented with the DCT and the
792:
267:
Moving objects within a frame are not sufficiently represented by global motion compensation. Thus, local
1566:
1561:
668:
407:
give the encoder the ability to dynamically choose what block size will be used to represent the motion.
162:
Differences between the original frame and the next frame, shifted right by 2 pixels. Shifting the frame
2624:
2508:
2435:
2273:
2253:
2197:
1855:
1646:
1449:
1377:
219:
309:
equivalent to a global motion compensation capable of panning. Further down the encoding pipeline, an
2609:
2518:
2459:
2385:
2233:
1823:
1818:
1673:
1516:
534:
526:
444:
294:
156:
59:
346:
frame into non-overlapping blocks, and the motion compensation vector tells where those blocks come
2523:
2096:
1890:
1591:
1361:
1296:
907:
596:
In 1977, Wen-Hsiung Chen developed a fast DCT algorithm with C.H. Smith and S.C. Fralick. In 1979,
285:
2464:
1835:
1722:
1678:
1491:
1474:
1464:
2089:
1840:
1624:
1469:
1313:
578:
55:
2361:
1050:
758:
2493:
1345:
1183:
1123:
718:
605:
461:
1252:
1217:
2177:
1639:
1601:
1422:
1185:
High
Definition Television: The Creation, Development and Implementation of HDTV Technology
1088:
951:
479:
compatible coding and can use motion compensation to compress between stereoscopic images.
8:
2299:
2258:
2243:
2212:
2207:
2116:
2023:
1956:
1910:
1693:
1352:
A New FFT Architecture and Chip Design for Motion
Compensation based on Phase Correlation
643:
488:
362:
precede or follow the predicted frame. These bidirectionally predicted frames are called
141:
1092:
955:
2481:
2451:
2430:
2336:
2268:
2162:
1850:
1666:
1656:
1551:
1531:
1526:
1351:
1104:
1031:
1013:
593:
scene with image quality comparable to an intra-frame coder requiring 2-bit per pixel.
542:
510:
126:
20:
2062:
2425:
2413:
2395:
2263:
2147:
2084:
1930:
1845:
1801:
1762:
1444:
1227:
1189:
1133:
1056:
963:
834:
825:
768:
728:
697:
680:
638:
558:
546:
538:
522:
328:
coefficients in high frequency sub-bands) due to quantization of coefficients of the
268:
102:
51:
39:
1108:
1035:
26:
2400:
2356:
2329:
2324:
2182:
2167:
2077:
1986:
1981:
1810:
1543:
1521:
1413:
1164:
1096:
1023:
986:
959:
764:
550:
530:
400:
333:
316:
It is possible to shift a block by a non-integer number of pixels, which is called
310:
167:
505:
researchers Y. Taki, M. Hatori and S. Tanaka, who proposed predictive inter-frame
16:
Video compression technique, used to efficiently predict and generate video frames
2319:
2133:
2057:
2038:
2008:
1976:
1942:
1501:
1439:
97:
2619:
2111:
1905:
1634:
1629:
1486:
1459:
1431:
1372:
1027:
590:
337:
1168:
990:
2603:
2418:
2366:
2033:
2028:
2003:
1935:
1556:
1454:
720:
Design of
Digital Video Coding Systems: A Complete Compressed Domain Approach
396:
255:
It does not partition the frames. This avoids artifacts at partition borders.
1004:
Chen, Z.; He, T.; Jin, X.; Wu, F. (2020). "Learning for Video
Compression".
838:
2539:
1506:
1481:
1382:
1257:
506:
494:
429:
235:
Pan — rotating the camera around its Y axis, moving the view left or right
2498:
2376:
2172:
2048:
1998:
648:
574:
569:
498:
238:
Tilt — rotating the camera around its X axis, moving the view up or down
2555:
2346:
2341:
2228:
2187:
1993:
694:
79:
1100:
105:, motion compensation) occupies more than 90% of encoding complexity.
724:
686:
653:
468:
457:
170:
of the camera, thus there is greater overlap between the two frames.
601:
coding technique for video compression from the late 1980s onwards.
2469:
2314:
1971:
1018:
933:
554:
2238:
1712:
1661:
663:
391:
video typically use a fixed block size, while newer ones such as
1378:
Compatibility between DCT, motion compensation and other methods
497:
scene that changed from frame-to-frame. In 1959, the concept of
1752:
1348:- article giving an overview of motion compensation techniques.
476:
388:
43:
1006:
IEEE Transactions on Circuits and Systems for Video Technology
222:, the motion model basically reflects camera motions such as:
2587:
2192:
1785:
1732:
874:
622:
617:
613:
609:
586:
472:
423:
392:
384:
63:
372:
capable of dealing with zooming, rotation, translation etc.
248:
There are several advantages of global motion compensation:
78:
Motion compensation exploits the fact that, often, for many
1742:
1596:
1581:
1571:
1219:
Standard Codecs: Image Compression to Advanced Video Coding
1087:. International Society for Optics and Photonics: 172–181.
626:
404:
321:
298:
182:
147:
Differences between the original frame and the next frame.
67:
689:
for playback of 24 frames per second movies on 60 Hz
375:
1717:
1683:
690:
582:
502:
368:. A coding scheme could, for instance, be IBBPBBPBBPBB.
977:
Habibi, Ali (1974). "Hybrid Coding of Pictorial Data".
430:
Quarter Pixel (QPel) and Half Pixel motion compensation
410:
245:
It works best for still scenes without moving objects.
1357:
DCT and DFT coefficients are related by simple factors
525:emerged with the development of motion-compensated
193:or bidirectionally from previous and future frames
938:"How I Came Up With the Discrete Cosine Transform"
2601:
1152:
62:(DCT). Most video coding standards, such as the
1253:"The History of Video File Formats Infographic"
1081:Efficient Transmission of Pictorial Information
241:Roll — rotating the camera around the view axis
1398:
1078:
438:
226:Dolly — moving the camera forward or backward
213:
1412:
827:Motion Compensated Video Coding - PhD Thesis
541:, and predictive motion compensation in the
279:
185:, images are predicted from previous frames
1211:
1209:
1207:
1205:
1003:
717:Chen, Jie; Koc, Ut-Va; Liu, KJ Ray (2001).
516:
1405:
1391:
1247:
1245:
1243:
716:
50:Motion compensation is one of the two key
1368:An Introduction to MPEG Video Compression
1332:Learn how and when to remove this message
1224:Institution of Engineering and Technology
1052:Multimedia Signal Coding and Transmission
1017:
342:Block motion compensation divides up the
274:
132:Full original frame, as shown on screen.
1295:This article includes a list of general
1215:
1202:
1074:
1072:
293:(BMC), also known as motion-compensated
25:
1240:
823:
381:Variable block-size motion compensation
376:Variable block-size motion compensation
229:Track — moving the camera left or right
2602:
1181:
976:
90:
1386:
1130:Springer Science & Business Media
1121:
1069:
932:
903:Video on demand: Research Paper 94/68
895:
893:
864:
862:
860:
858:
856:
854:
852:
850:
848:
553:technique that was first proposed by
1281:
926:
752:
750:
748:
746:
744:
501:motion compensation was proposed by
416:Overlapped block motion compensation
411:Overlapped block motion compensation
350:(a common misconception is that the
1156:IEEE Transactions on Communications
1048:
979:IEEE Transactions on Communications
899:
791:. February 20, 2009. Archived from
443:Motion compensation is utilized in
232:Boom — moving the camera up or down
42:, for example in the generation of
13:
1362:DCT better than DFT also for video
1301:it lacks sufficient corresponding
890:
845:
756:
14:
2641:
1277:
741:
566:University of Southern California
426:Annex F Advanced Prediction mode
2579:
2578:
2569:
2568:
1286:
557:, who initially intended it for
435:process" of the H.264 standard.
155:
140:
125:
73:
1175:
1146:
1115:
1042:
997:
970:
674:
659:Television standards conversion
870:"History of Video Compression"
817:
808:
799:
781:
710:
152:Motion compensated difference
1:
703:
521:Practical motion-compensated
964:10.1016/1051-2004(91)90086-Z
833:. University of Nottingham.
471:Encoding techniques utilize
47:efficiency can be improved.
7:
1373:DCT and motion compensation
1216:Ghanbari, Mohammed (2003).
669:X-Video Motion Compensation
632:
564:In 1974, Ali Habibi at the
10:
2646:
2460:Compressed data structures
1782:RLE + BWT + MTF + Huffman
1450:Asymmetric numeral systems
1182:Cianci, Philip J. (2014).
1028:10.1109/TCSVT.2019.2892608
824:Garnham, Nigel W. (1995).
486:
482:
439:3D image coding techniques
283:
220:global motion compensation
214:Global motion compensation
18:
2630:Motion in computer vision
2615:Film and video technology
2564:
2548:
2532:
2450:
2375:
2307:
2298:
2221:
2155:
2146:
2047:
1964:
1955:
1871:
1819:Discrete cosine transform
1809:
1800:
1749:LZ77 + Huffman + context
1702:
1612:
1542:
1430:
1421:
1188:. McFarland. p. 63.
1169:10.1109/TCOM.1977.1093941
1055:. Springer. p. 364.
1049:Ohm, Jens-Rainer (2015).
991:10.1109/TCOM.1974.1092258
943:Digital Signal Processing
535:discrete cosine transform
445:stereoscopic video coding
330:Fourier-related transform
295:discrete cosine transform
291:Block motion compensation
280:Block motion compensation
60:discrete cosine transform
2524:Smallest grammar problem
1346:Temporal Rate Conversion
908:House of Commons Library
629:formats) that followed.
286:Block-matching algorithm
19:Not to be confused with
2465:Compressed suffix array
2014:Nyquist–Shannon theorem
1316:more precise citations.
1125:Image Sequence Analysis
176:
757:Li, Jian Ping (2006).
579:fast Fourier transform
517:Motion-compensated DCT
462:affine transformations
275:Motion-compensated DCT
56:video coding standards
31:
2494:Kolmogorov complexity
2362:Video characteristics
1739:LZ77 + Huffman + ANS
1122:Huang, T. S. (1981).
900:Lea, William (1994).
606:video coding standard
29:
2584:Compression software
2178:Compression artifact
2134:Psychoacoustic model
914:on 20 September 2019
537:(DCT) coding in the
2574:Compression formats
2213:Texture compression
2208:Standard test image
2024:Silence compression
1093:1975SPIE...66..172R
956:1991DSP.....1....4A
644:Image stabilization
612:, developed by the
489:Video coding format
318:sub-pixel precision
91:Illustrated example
54:techniques used in
35:Motion compensation
2482:Information theory
2337:Display resolution
2163:Chroma subsampling
1552:Byte pair encoding
1497:Shannon–Fano–Elias
604:The first digital
549:block compression
545:. DCT coding is a
543:temporal dimension
511:temporal dimension
32:
21:motion compensator
2625:Video compression
2597:
2596:
2446:
2445:
2396:Deblocking filter
2294:
2293:
2142:
2141:
1951:
1950:
1796:
1795:
1342:
1341:
1334:
1101:10.1117/12.965361
698:cathode ray tubes
681:video compression
639:Motion estimation
559:image compression
539:spatial dimension
523:video compression
269:motion estimation
174:
173:
103:motion estimation
58:, along with the
52:video compression
40:video compression
2637:
2610:Data compression
2582:
2581:
2572:
2571:
2401:Lapped transform
2305:
2304:
2183:Image resolution
2168:Coding tree unit
2153:
2152:
1962:
1961:
1807:
1806:
1428:
1427:
1414:Data compression
1407:
1400:
1393:
1384:
1383:
1337:
1330:
1326:
1323:
1317:
1312:this article by
1303:inline citations
1290:
1289:
1282:
1271:
1270:
1268:
1266:
1249:
1238:
1237:
1226:. pp. 1–2.
1213:
1200:
1199:
1179:
1173:
1172:
1163:(9): 1004–1009.
1150:
1144:
1143:
1119:
1113:
1112:
1076:
1067:
1066:
1046:
1040:
1039:
1021:
1001:
995:
994:
974:
968:
967:
936:(January 1991).
930:
924:
923:
921:
919:
910:. Archived from
897:
888:
887:
885:
883:
866:
843:
842:
832:
821:
815:
812:
806:
803:
797:
796:
785:
779:
778:
765:World Scientific
754:
739:
738:
714:
551:transform coding
531:data compression
401:H.264/MPEG-4 AVC
334:transform coding
271:is also needed.
207:
203:
200:
192:
159:
144:
129:
108:
107:
2645:
2644:
2640:
2639:
2638:
2636:
2635:
2634:
2600:
2599:
2598:
2593:
2560:
2544:
2528:
2509:Rate–distortion
2442:
2371:
2290:
2217:
2138:
2043:
2039:Sub-band coding
1947:
1872:Predictive type
1867:
1792:
1759:LZSS + Huffman
1709:LZ77 + Huffman
1698:
1608:
1544:Dictionary type
1538:
1440:Adaptive coding
1417:
1411:
1338:
1327:
1321:
1318:
1308:Please help to
1307:
1291:
1287:
1280:
1275:
1274:
1264:
1262:
1261:. 22 April 2012
1251:
1250:
1241:
1234:
1214:
1203:
1196:
1180:
1176:
1151:
1147:
1140:
1120:
1116:
1077:
1070:
1063:
1047:
1043:
1002:
998:
975:
971:
931:
927:
917:
915:
898:
891:
881:
879:
868:
867:
846:
830:
822:
818:
813:
809:
804:
800:
787:
786:
782:
775:
767:. p. 847.
755:
742:
735:
715:
711:
706:
693:or 100 Hz
677:
635:
519:
491:
485:
441:
432:
413:
378:
338:residual frames
288:
282:
277:
216:
205:
201:
194:
186:
179:
98:Elephants Dream
93:
76:
24:
17:
12:
11:
5:
2643:
2633:
2632:
2627:
2622:
2617:
2612:
2595:
2594:
2592:
2591:
2576:
2565:
2562:
2561:
2559:
2558:
2552:
2550:
2546:
2545:
2543:
2542:
2536:
2534:
2530:
2529:
2527:
2526:
2521:
2516:
2511:
2506:
2501:
2496:
2491:
2490:
2489:
2479:
2474:
2473:
2472:
2467:
2456:
2454:
2448:
2447:
2444:
2443:
2441:
2440:
2439:
2438:
2433:
2423:
2422:
2421:
2416:
2411:
2403:
2398:
2393:
2388:
2382:
2380:
2373:
2372:
2370:
2369:
2364:
2359:
2354:
2349:
2344:
2339:
2334:
2333:
2332:
2327:
2322:
2311:
2309:
2302:
2296:
2295:
2292:
2291:
2289:
2288:
2287:
2286:
2281:
2276:
2271:
2261:
2256:
2251:
2246:
2241:
2236:
2231:
2225:
2223:
2219:
2218:
2216:
2215:
2210:
2205:
2200:
2195:
2190:
2185:
2180:
2175:
2170:
2165:
2159:
2157:
2150:
2144:
2143:
2140:
2139:
2137:
2136:
2131:
2126:
2125:
2124:
2119:
2114:
2109:
2104:
2094:
2093:
2092:
2082:
2081:
2080:
2075:
2065:
2060:
2054:
2052:
2045:
2044:
2042:
2041:
2036:
2031:
2026:
2021:
2016:
2011:
2006:
2001:
1996:
1991:
1990:
1989:
1984:
1979:
1968:
1966:
1959:
1953:
1952:
1949:
1948:
1946:
1945:
1943:Psychoacoustic
1940:
1939:
1938:
1933:
1928:
1920:
1919:
1918:
1913:
1908:
1903:
1898:
1888:
1887:
1886:
1875:
1873:
1869:
1868:
1866:
1865:
1864:
1863:
1858:
1853:
1843:
1838:
1833:
1832:
1831:
1826:
1815:
1813:
1811:Transform type
1804:
1798:
1797:
1794:
1793:
1791:
1790:
1789:
1788:
1780:
1779:
1778:
1775:
1767:
1766:
1765:
1757:
1756:
1755:
1747:
1746:
1745:
1737:
1736:
1735:
1727:
1726:
1725:
1720:
1715:
1706:
1704:
1700:
1699:
1697:
1696:
1691:
1686:
1681:
1676:
1671:
1670:
1669:
1664:
1654:
1649:
1644:
1643:
1642:
1632:
1627:
1622:
1616:
1614:
1610:
1609:
1607:
1606:
1605:
1604:
1599:
1594:
1589:
1584:
1579:
1574:
1569:
1564:
1554:
1548:
1546:
1540:
1539:
1537:
1536:
1535:
1534:
1529:
1524:
1519:
1509:
1504:
1499:
1494:
1489:
1484:
1479:
1478:
1477:
1472:
1467:
1457:
1452:
1447:
1442:
1436:
1434:
1425:
1419:
1418:
1410:
1409:
1402:
1395:
1387:
1381:
1380:
1375:
1370:
1366:John Wiseman,
1364:
1359:
1354:
1349:
1340:
1339:
1294:
1292:
1285:
1279:
1278:External links
1276:
1273:
1272:
1239:
1232:
1201:
1194:
1174:
1145:
1138:
1132:. p. 29.
1114:
1068:
1061:
1041:
1012:(2): 566–576.
996:
985:(5): 614–624.
969:
925:
889:
844:
816:
807:
798:
795:on 2009-02-20.
780:
773:
740:
733:
727:. p. 71.
708:
707:
705:
702:
701:
700:
683:
676:
673:
672:
671:
666:
661:
656:
651:
646:
641:
634:
631:
591:videotelephone
518:
515:
487:Main article:
484:
481:
440:
437:
431:
428:
412:
409:
377:
374:
352:previous frame
281:
278:
276:
273:
264:compensation.
261:
260:
256:
253:
243:
242:
239:
236:
233:
230:
227:
215:
212:
178:
175:
172:
171:
160:
153:
149:
148:
145:
138:
134:
133:
130:
123:
119:
118:
115:
114:Example Frame
112:
92:
89:
75:
72:
15:
9:
6:
4:
3:
2:
2642:
2631:
2628:
2626:
2623:
2621:
2618:
2616:
2613:
2611:
2608:
2607:
2605:
2589:
2585:
2577:
2575:
2567:
2566:
2563:
2557:
2554:
2553:
2551:
2547:
2541:
2538:
2537:
2535:
2531:
2525:
2522:
2520:
2517:
2515:
2512:
2510:
2507:
2505:
2502:
2500:
2497:
2495:
2492:
2488:
2485:
2484:
2483:
2480:
2478:
2475:
2471:
2468:
2466:
2463:
2462:
2461:
2458:
2457:
2455:
2453:
2449:
2437:
2434:
2432:
2429:
2428:
2427:
2424:
2420:
2417:
2415:
2412:
2410:
2407:
2406:
2404:
2402:
2399:
2397:
2394:
2392:
2389:
2387:
2384:
2383:
2381:
2378:
2374:
2368:
2367:Video quality
2365:
2363:
2360:
2358:
2355:
2353:
2350:
2348:
2345:
2343:
2340:
2338:
2335:
2331:
2328:
2326:
2323:
2321:
2318:
2317:
2316:
2313:
2312:
2310:
2306:
2303:
2301:
2297:
2285:
2282:
2280:
2277:
2275:
2272:
2270:
2267:
2266:
2265:
2262:
2260:
2257:
2255:
2252:
2250:
2247:
2245:
2242:
2240:
2237:
2235:
2232:
2230:
2227:
2226:
2224:
2220:
2214:
2211:
2209:
2206:
2204:
2201:
2199:
2196:
2194:
2191:
2189:
2186:
2184:
2181:
2179:
2176:
2174:
2171:
2169:
2166:
2164:
2161:
2160:
2158:
2154:
2151:
2149:
2145:
2135:
2132:
2130:
2127:
2123:
2120:
2118:
2115:
2113:
2110:
2108:
2105:
2103:
2100:
2099:
2098:
2095:
2091:
2088:
2087:
2086:
2083:
2079:
2076:
2074:
2071:
2070:
2069:
2066:
2064:
2061:
2059:
2056:
2055:
2053:
2050:
2046:
2040:
2037:
2035:
2034:Speech coding
2032:
2030:
2029:Sound quality
2027:
2025:
2022:
2020:
2017:
2015:
2012:
2010:
2007:
2005:
2004:Dynamic range
2002:
2000:
1997:
1995:
1992:
1988:
1985:
1983:
1980:
1978:
1975:
1974:
1973:
1970:
1969:
1967:
1963:
1960:
1958:
1954:
1944:
1941:
1937:
1934:
1932:
1929:
1927:
1924:
1923:
1921:
1917:
1914:
1912:
1909:
1907:
1904:
1902:
1899:
1897:
1894:
1893:
1892:
1889:
1885:
1882:
1881:
1880:
1877:
1876:
1874:
1870:
1862:
1859:
1857:
1854:
1852:
1849:
1848:
1847:
1844:
1842:
1839:
1837:
1834:
1830:
1827:
1825:
1822:
1821:
1820:
1817:
1816:
1814:
1812:
1808:
1805:
1803:
1799:
1787:
1784:
1783:
1781:
1776:
1774:
1771:
1770:
1769:LZ77 + Range
1768:
1764:
1761:
1760:
1758:
1754:
1751:
1750:
1748:
1744:
1741:
1740:
1738:
1734:
1731:
1730:
1728:
1724:
1721:
1719:
1716:
1714:
1711:
1710:
1708:
1707:
1705:
1701:
1695:
1692:
1690:
1687:
1685:
1682:
1680:
1677:
1675:
1672:
1668:
1665:
1663:
1660:
1659:
1658:
1655:
1653:
1650:
1648:
1645:
1641:
1638:
1637:
1636:
1633:
1631:
1628:
1626:
1623:
1621:
1618:
1617:
1615:
1611:
1603:
1600:
1598:
1595:
1593:
1590:
1588:
1585:
1583:
1580:
1578:
1575:
1573:
1570:
1568:
1565:
1563:
1560:
1559:
1558:
1555:
1553:
1550:
1549:
1547:
1545:
1541:
1533:
1530:
1528:
1525:
1523:
1520:
1518:
1515:
1514:
1513:
1510:
1508:
1505:
1503:
1500:
1498:
1495:
1493:
1490:
1488:
1485:
1483:
1480:
1476:
1473:
1471:
1468:
1466:
1463:
1462:
1461:
1458:
1456:
1453:
1451:
1448:
1446:
1443:
1441:
1438:
1437:
1435:
1433:
1429:
1426:
1424:
1420:
1415:
1408:
1403:
1401:
1396:
1394:
1389:
1388:
1385:
1379:
1376:
1374:
1371:
1369:
1365:
1363:
1360:
1358:
1355:
1353:
1350:
1347:
1344:
1343:
1336:
1333:
1325:
1315:
1311:
1305:
1304:
1298:
1293:
1284:
1283:
1260:
1259:
1254:
1248:
1246:
1244:
1235:
1233:9780852967102
1229:
1225:
1221:
1220:
1212:
1210:
1208:
1206:
1197:
1195:9780786487974
1191:
1187:
1186:
1178:
1170:
1166:
1162:
1158:
1157:
1149:
1141:
1139:9783642870378
1135:
1131:
1127:
1126:
1118:
1110:
1106:
1102:
1098:
1094:
1090:
1086:
1082:
1075:
1073:
1064:
1062:9783662466919
1058:
1054:
1053:
1045:
1037:
1033:
1029:
1025:
1020:
1015:
1011:
1007:
1000:
992:
988:
984:
980:
973:
965:
961:
957:
953:
949:
945:
944:
939:
935:
929:
913:
909:
905:
904:
896:
894:
877:
876:
871:
865:
863:
861:
859:
857:
855:
853:
851:
849:
840:
836:
829:
828:
820:
811:
802:
794:
790:
784:
776:
774:9789812709998
770:
766:
762:
761:
753:
751:
749:
747:
745:
736:
734:9780203904183
730:
726:
722:
721:
713:
709:
699:
696:
692:
688:
684:
682:
679:
678:
670:
667:
665:
662:
660:
657:
655:
652:
650:
647:
645:
642:
640:
637:
636:
630:
628:
624:
619:
615:
611:
607:
602:
599:
594:
592:
588:
584:
580:
576:
571:
567:
562:
560:
556:
552:
548:
544:
540:
536:
532:
528:
524:
514:
512:
508:
504:
500:
496:
490:
480:
478:
474:
470:
466:
463:
459:
455:
453:
448:
446:
436:
427:
425:
419:
417:
408:
406:
402:
398:
397:MPEG-4 Part 2
394:
390:
386:
382:
373:
369:
367:
366:
359:
357:
353:
349:
345:
340:
339:
335:
331:
325:
323:
319:
314:
312:
311:entropy coder
306:
304:
303:motion vector
300:
296:
292:
287:
272:
270:
265:
257:
254:
251:
250:
249:
246:
240:
237:
234:
231:
228:
225:
224:
223:
221:
211:
208:
198:
190:
184:
169:
165:
161:
158:
154:
151:
150:
146:
143:
139:
136:
135:
131:
128:
124:
121:
120:
116:
113:
110:
109:
106:
104:
100:
99:
88:
84:
81:
74:Functionality
71:
69:
65:
61:
57:
53:
48:
45:
41:
36:
28:
22:
2540:Hutter Prize
2504:Quantization
2409:Compensation
2408:
2203:Quantization
1926:Compensation
1925:
1492:Shannon–Fano
1432:Entropy type
1328:
1322:October 2013
1319:
1300:
1263:. Retrieved
1258:RealNetworks
1256:
1218:
1184:
1177:
1160:
1154:
1148:
1124:
1117:
1084:
1080:
1051:
1044:
1009:
1005:
999:
982:
978:
972:
947:
941:
934:Ahmed, Nasir
928:
918:20 September
916:. Retrieved
912:the original
902:
880:. Retrieved
873:
826:
819:
810:
801:
793:the original
789:"MPEG-2 FAQ"
783:
759:
719:
712:
675:Applications
603:
598:Anil K. Jain
595:
563:
533:techniques:
520:
507:video coding
495:analog video
492:
467:
456:
451:
449:
442:
433:
420:
415:
414:
380:
379:
370:
364:
360:
355:
351:
347:
343:
341:
326:
317:
315:
307:
302:
290:
289:
266:
262:
247:
244:
217:
209:
180:
163:
117:Description
96:
94:
85:
77:
49:
34:
33:
2499:Prefix code
2352:Frame types
2173:Color space
1999:Convolution
1729:LZ77 + ANS
1640:Incremental
1613:Other types
1532:Levenshtein
1314:introducing
649:Inter frame
575:inter-frame
570:intra-frame
561:, in 1972.
555:Nasir Ahmed
499:inter-frame
164:compensates
137:Difference
2604:Categories
2556:Mark Adler
2514:Redundancy
2431:Daubechies
2414:Estimation
2347:Frame rate
2269:Daubechies
2229:Chain code
2188:Macroblock
1994:Companding
1931:Estimation
1851:Daubechies
1557:Lempel–Ziv
1517:Exp-Golomb
1445:Arithmetic
1297:references
1019:1804.09869
950:(1): 4–5.
882:3 November
704:References
695:interlaced
685:change of
450:In video,
284:See also:
259:direction.
2533:Community
2357:Interlace
1743:Zstandard
1522:Fibonacci
1512:Universal
1470:Canonical
725:CRC Press
687:framerate
654:HDTV blur
458:JPEG 2000
332:used for
206:B frames.
122:Original
2519:Symmetry
2487:Timeline
2470:FM-index
2315:Bit rate
2308:Concepts
2156:Concepts
2019:Sampling
1972:Bit rate
1965:Concepts
1667:Sequitur
1502:Tunstall
1475:Modified
1465:Adaptive
1423:Lossless
1265:5 August
1109:62725808
1036:13743007
839:59633188
633:See also
469:2D+Delta
365:B-frames
202:B frames
197:B frames
189:P frames
166:for the
2477:Entropy
2426:Wavelet
2405:Motion
2264:Wavelet
2244:Fractal
2239:Deflate
2222:Methods
2009:Latency
1922:Motion
1846:Wavelet
1763:LHA/LZH
1713:Deflate
1662:Re-Pair
1657:Grammar
1487:Shannon
1460:Huffman
1416:methods
1310:improve
1089:Bibcode
952:Bibcode
664:VidFIRE
509:in the
483:History
344:current
336:of the
168:panning
2588:codecs
2549:People
2452:Theory
2419:Vector
1936:Vector
1753:Brotli
1703:Hybrid
1602:Snappy
1455:Golomb
1299:, but
1230:
1192:
1136:
1107:
1059:
1034:
837:
771:
731:
589:for a
477:MPEG-2
403:, and
389:MPEG-1
80:frames
44:MPEG-2
2620:H.26x
2379:parts
2377:Codec
2342:Frame
2300:Video
2284:SPIHT
2193:Pixel
2148:Image
2102:ACELP
2073:ADPCM
2063:ÎĽ-law
2058:A-law
2051:parts
2049:Codec
1957:Audio
1896:ACELP
1884:ADPCM
1861:SPIHT
1802:Lossy
1786:bzip2
1777:LZHAM
1733:LZFSE
1635:Delta
1527:Gamma
1507:Unary
1482:Range
1105:S2CID
1032:S2CID
1014:arXiv
875:ITU-T
831:(PDF)
623:H.26x
618:H.261
614:CCITT
610:H.120
587:pixel
547:lossy
473:H.264
424:H.263
393:H.263
385:H.261
111:Type
64:H.26x
2391:DPCM
2198:PSNR
2129:MDCT
2122:WLPC
2107:CELP
2068:DPCM
1916:WLPC
1901:CELP
1879:DPCM
1829:MDCT
1773:LZMA
1674:LDCT
1652:DPCM
1597:LZWL
1587:LZSS
1582:LZRW
1572:LZJB
1267:2019
1228:ISBN
1190:ISBN
1134:ISBN
1085:0066
1057:ISBN
920:2019
884:2019
835:OCLC
769:ISBN
729:ISBN
691:LCDs
627:MPEG
625:and
608:was
585:per
475:and
452:time
405:VC-1
387:and
348:from
322:Qpel
299:MPEG
183:MPEG
177:MPEG
68:MPEG
66:and
2436:DWT
2386:DCT
2330:VBR
2325:CBR
2320:ABR
2279:EZW
2274:DWT
2259:RLE
2249:KLT
2234:DCT
2117:LSP
2112:LAR
2097:LPC
2090:FFT
1987:VBR
1982:CBR
1977:ABR
1911:LSP
1906:LAR
1891:LPC
1856:DWT
1841:FFT
1836:DST
1824:DCT
1723:LZS
1718:LZX
1694:RLE
1689:PPM
1684:PAQ
1679:MTF
1647:DMC
1625:CTW
1620:BWT
1592:LZW
1577:LZO
1567:LZ4
1562:842
1165:doi
1097:doi
1024:doi
987:doi
960:doi
583:bit
527:DCT
503:NHK
218:In
181:In
2606::
2254:LP
2085:FT
2078:DM
1630:CM
1255:.
1242:^
1222:.
1204:^
1161:25
1159:.
1128:.
1103:.
1095:.
1083:.
1071:^
1030:.
1022:.
1010:30
1008:.
983:22
981:.
958:.
946:.
940:.
906:.
892:^
872:.
847:^
763:.
743:^
723:.
513:.
447:.
399:,
395:,
356:to
305:.
199:).
2590:)
2586:(
1406:e
1399:t
1392:v
1335:)
1329:(
1324:)
1320:(
1306:.
1269:.
1236:.
1198:.
1171:.
1167::
1142:.
1111:.
1099::
1091::
1065:.
1038:.
1026::
1016::
993:.
989::
966:.
962::
954::
948:1
922:.
886:.
841:.
777:.
737:.
195:(
191:)
187:(
23:.
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.
