"H.264 explain the type and level
Transfer from http://blog.csdn.net/nonmarking/article/details/21242689
H.264 There are four quality levels, namely BP, EP, MP, HP:
1, BP-Baseline Profile: basic quality. Support I / P frame, supports only non interlaced (Progressive) and CAVLC; 2, EP-Extended profile: Advanced quality. Support I / P / B / SP / SI-frame, supports only non-interlaced (Progressive) and CAVLC; 3, MP-Main profile: main picture quality. Providing I / P / B-frame, non interlaced support (Progressive) and interleaved (Interlaced), also supports CABAC and CAVLC is supported; 4, HP-High profile: High quality. On the basis of the main Profile adds 8x8 intra-prediction, custom quantization, lossless video coding and more YUV format;
H.264 HP wants to illustrate the difference with H.264 MP talked about the necessary technological development of the H.264. JVT completed in 2003 H.264 essential part of the standard-setting work, include Baseline profile, Extended profile and Main profile, include different coding tools. After they completed the JVT H.264 FRExt (ie: Fidelity Range Extensions) extensions (Amendment) development work, including High profile (HP), High 10 profile (Hi10P), High 4: 2: 2 profile (Hi422P), High 4: 4: 4 profile (Hi444P) 4 Ge profile.
H.264 Baseline profile, Extended profile Main profile and sample data are for 8, 4: 2: 0 format video sequence, extending it to FRExt 8 to 12 sample data, the video format is 4: 2: 0 , 4: 2: 2 and 4: 4: 4, set up a High profile (HP), High 10 profile (Hi10P), High 4: 2: 2 profile (Hi422P), High 4: 4: 4 profile (Hi444P) 4 a profile, this profile are four to Main profile basis.
In the same configuration, High profile (HP) savings ratio Main profile (MP) 10% of the code flow, than the MPEG-2 MP save 60% of the traffic code has better coding performance. According to different application areas, Baseline profile and more used in real-time communications, Main profile and more used in streaming media, High profile is more used in broadcasting and storage areas.
H.264 Baseline Profile corresponding to the MPEG-4 SP H.264 Main Profile corresponding to the MPEG-4 ASP H.264 Extended Profile corresponding to the MPEG-4 ARTS or FGS H.264 Baseline Profile corresponding to the MPEG-4 Studio.
Be
The following diagram is a reflection of the relationship between the three H.264 Profile, Level, Encoder's
h.264 profile
Level (Level) is used to constrain the resolution, frame rate and bit rate. For more information, see the table -
Max macroblocks: the maximum number of macro blocks. Note: the macroblock size is 16x16.
per second: (the maximum number of macro blocks) per second. It can be used to constrain the frame rate.
per frame: each frame (the maximum number of macro blocks). It can be used to constrain the resolution.
Max video bit rate (kbit / s): The maximum video bit rate. It will be different under different grades (Profile).
BP: Baseline Profile, Baseline Profile.
XP: Extended Profile, advanced grade.
MP: Main Profile, Main Profile.
HiP: High Profile, senior grade.
Hi10P: High 10 Profile, High 10 grade.
Hi422P: High 4: 2: 2 Profile, High 4: 2: 2 level.
Hi444PP: High 4: 4: 4 Predictive Profile, High 4: 4: 4 (? Experimental) grades.
Examples for high resolution @ frame rate (max stored frames): Example: @ high resolution frame rate (the maximum storage number of frames). h.264 software encoder
For H.264 (MPEG-4 AVC), level (Level) with grade (Profile
) as important parameters. But a lot of articles that is very simple, just say the general standard definition video with 3.1, with 4 HD, specific meaning did not elaborate.
In video compression, bit rate selection is very easy to judge. The relationship between the resolution and frame rate and level, not so obvious. So we need to find the method of calculation. Be
First look at the video resolution of the common supporting level -
480P (800x480): When the frame rate is lower than 13.5 with 2.2, 3, if the frame rate to use between 13.5 to 27, 27 with a frame rate higher than 3.1.
720P (1280x720): General 3.1.
1080P (1920x1080): usually 4.
We can see 480P (800x480) a little special, so I have it as an example to explain how to calculate the resolution and frame rate.
Due to the macroblock size is 16x16, and based on this information we calculate the horizontal macroblocks (PicWidthInMbs) and vertical macroblocks (FrameHeightInMbs) -
The number of horizontal macro blocks = ceil (video width / 16) = ceil (800/16) = ceil (50.0) = 50
Vertical macroblocks = ceil (video height / 16) = ceil (480/16) = ceil (30.0) = 30
NOTE: ceil (x) is the function of the top round, it returns the smallest integer greater than or equal to x.
Be
It is then calculated for each frame macroblocks (macroblocks per frame) -
Of macroblocks per frame = number of macroblocks in the horizontal * vertical macroblocks = 1500 = 50 * 30
Be
The above detailed table search level, the lowest level seen per support frame macroblock 1500 is 2.2.
The maximum number of 2.2 per macroblock level allowed (Max macroblocks per second) is 20250.20250 / 1500 = 13.5. That is up to 13.5 frames per second.
One level, the maximum number of macroblocks per second level 3 is allowed 40500.40500 / 27 = 1500. That is up to 27 per second.
Then one level, the maximum number of macroblocks per second level allowed is 108000.108000 3.1 / 1500 = 72. That is up to 72 frames per second, enough.
Be
When a macro block exceeds level constraint, x264 displays such a warning message -
x264 [warning]: frame MB size (120x68)> level limit (5120)
Level detailed in the preceding table, there is a mysterious field - the maximum number of frames stored (max stored frames). It's come from it?
Be
DPB maximum storage frames and related --DPB: decoded picture buffer, the decoded picture buffer. When do video decoding, you need a number of recent pieces of reference frame cached, this buffer is called the DPB. Therefore, the maximum number of frames stored in the reference frame number is the largest (ref). DPB generally in units of macroblocks (DpbMbs), calculated as -
DpbMbs = ref (reference frames) * PicWidthInMbs (horizontal macroblock number) * FrameHeightInMbs (vertical macroblocks)
Be
At different levels (Level), the largest of the decoded picture buffer macro blocks (MaxDpbMbs) is different -
Level MaxDpbMbs 1 396 1b 396 1.1 900 1.2 2,376 1.3 2,376 2 2,376 2.1 4,752 2.2 8,100 3 8,100 3.1 18,000 3.2 20,480 4 32,768 4.1 32,768 4.2 34,816 5 110,400 5.1 184,320
Be
We can pour the launch according to MaxDpbMbs restricted maximum number of frames of reference -
max_ref = min (floor (MaxDpbMbs / (PicWidthInMbs * FrameHeightInMbs)), 16)
Note 1: floor (x) is a function of the end of the round, it returns the largest integer less than or equal to x.
Note 2: the number of frames by reference (ref) only up to 16.
Reprinted address: http: //www.cnblogs.com/zyl910/archive/2011/12/08/h264_level.html
http://xcshen.blog.51cto.com/2835389/552669
================================================== ================================================== =====
MP has a relationship with B-frames. AVC / H.264 defines four different Profile (class): Baseline, Main, Extended
Transfer from http://www.hzlitai.com.cn/article/ARM11/CPhelp/1772.html
H.264 standard three frames
H.264 codec standard previous frame as set forth in H.261, H.263 and MPEG-1/2/4 did not change significantly, also based on hybrid coding schemes: vectors representing the motion image sequence to each frame motion content, using its previously decoded frame motion estimation and compensation or intra prediction techniques, the resulting mixed image to the converted value, quantization, entropy coding processing section. Therefore, new standards and improving performance is to apply the new algorithm aspect of the various parts. A lot of work in the new standard to improve the fault tolerance of image transmission, is adapted to redefine the configuration of the image division. When coding, each video frame is divided into a plurality of portions of the structure to slice, each slice can be decoded independently of other parts are not affected. Slice the basic structure of the image - macroblocks, each macroblock comprising 16 × 16 block of luminance and two chrominance blocks of 8 × 8. To further improve the robustness of the whole
The system is divided into
And a video coding layer network abstraction layer. Video coding layer describes video data to be transmitted is carried
Video content. And the network abstraction layer is considered different applications, such as video transmission or video RTP communication session, H.32X consecutive packets / UDP / IP communication. Be
The H.264 standard is divided into three frames (profile): Baseline, Main and X, it represents a defined set of algorithms for different applications and technologies. Wherein, Baseline mainly comprising a low-complexity, low-latency technical features; mainly for interactive applications; fault tolerance into consideration, Baseline content harsh environment are substantially higher level of the other profile included. The Main profile is higher coding efficiency for applications such as video broadcasting. X profile is designed primarily for streaming media applications, in this framework, all fault-tolerant technology and flexible access technology handover and bitstream are included.
The main technical features of the Baseline profile
⑴ Baseline decoder only for I slice and the P slice operation. For inter prediction, compared to the previous standard
Registration, in order to more accurately predict the motion compensation picture content, the new standard allows further divided into macroblocks 16 × 16,16 × 8,8 × 16,8 × 8,8 × 4,4 × 8,4 × 4 sub-blocks; a motion estimation accuracy of 1/4 pixel position obtained via the 6-tap filter; neighboring block motion vectors obtained by the prediction, the prediction difference is encoded transmission. H.264 supports multiple reference frame prediction, motion estimation reference frames specified for use up to 15, the use of multiple reference frames greatly improving the fault tolerance of image transmission, suppressionThe error in space and
Spread over time. For all types of slice encoding, H.264 supports two types of intra-coded: 4x4 and 16x16 coding mode. For the 4x4 mode each 4x4 luminance block of eight kinds of prediction modes in different directions, and a DC prediction mode. For the 16x16 mode, each luminance block has four 16x16 intra prediction modes. For 8x8 chrominance macroblock samples, using almost the same luminance 16x16 prediction mode. In order to ensure the independence of slice of coding, intra prediction is not allowed across slice boundaries. For transformation, quantization section. Unlike the previous transform coding standard using prediction values of DCT mixed, H.264 uses a simple integer transformation. This transformation is almost the same as compared with the DCT compressing performance and has many advantages, its core shifting of only addition and subtraction, shift operation, to avoid loss of precision. Quantize the transform coefficients using the stagger step size quantizer 52, while only 31 H.263 standard. Quantization step size increments of 12.5%, to expand the range of the quantization step-like encoder more flexible and precise control, between the bit rate and the image quality to achieve a compromise. Entropy coding part. For the quantized transform coefficients to be transmitted, if the context of variable length coding (CAVLC) on which coded coefficients are selected to be used the next variable-length coding the quantized transform coefficient table in front of the size of the transmission in accordance with the value coded. Since variable-length coding table is designed based on the respective statistical conditions, the performance is better than using a single variable length coding table. While other data such as header information and the like, the use of a single variable-length coding table (Exp-Golomb code). New standards are still based prediction blocks and reconstructed embodiment, in order to remove the blocking effect of the impact image resulting subjective quality, H.264 uses a deblocking filter. The main idea is that when a small difference on both sides of the block boundary filter is used so that the difference "smooth" out, if the image feature boundary filtering is not used obviously. This will not only weaken as the "block effect" and avoid the effects of the objective characteristics filtered image. At the same time so that the bit rate of 5-10% in the same subjective quality. Be Further, the organization and transfer of the image data. Macroblock picture in the H.264 standard may be flexible macroblock order organization (FMO) is divided into a plurality of slice Group; slice is transmitted between each independently may be any order to the decoding side (ASO). Slice in the bitstream and may be used a repeated manner (RS) transmissions, which can be used to restore the data in case of error slice, to enhance the robustness of the transmission of images. While the independence between the slice error spatial propagation is suppressed, thus improving the fault tolerance of the bit stream. Technical characteristics of ⑵ Main profile Main profile contains all the algorithms Baseline profile and has additional technical features, but it does not support FMO, ASO and RS technologies. Processing operations on the support only I, P, B slice of. Within this framework proposed the concept of adaptation transform block division size (of ABT) is. This concept is for inter-coded, the main idea is to predict will be mixed transform coding block size linked to the block size used for motion compensation. This use of the largest possible
Transform coding the signal length. However, due to the complexity, the maximum block size is converted is limited to 8 × 8 or less. Entropy coding section, a more efficient encoding, as used herein, context-based arithmetic coding (CABAC) so that entropy coding performance is further improved. Compared CAVLC, CABAC coded television signal used in the same picture quality will reduce the bit rate of 10-15%. Another, Main profile division multiple slice group is not supported.
⑶ related coding How issues key research project selection (mode decision) and using motion estimation method (ME) has always realize the video encoding prediction model has been proposed. In the implementation of the H.263 standard
Software selection model is based on a simple comparison of the threshold. Standard test used in the new software in the Lagrangian rate-distortion optimization strategy, which is based on a code rate used for each image block size and the prediction mode is generated for each mixed and transmitted. In this way, the mode selection can achieve optimal rate - distortion performance increase but at the expense of computational complexity. This optimization is to minimize the following Lagrangian pull function: J = SATD + λ R wherein, R is the transmission bit rate corresponding to each portion; [lambda] parameter to optimize, which has a strong correlation with the quantization parameter? . SATD is predicted through a 4 × 4 Hadamard transform block sum of absolute values mixed. For all intra, inter macroblock coding modes and reference frames are selected plurality of minimizing the Lagrangian function is achieved by pulling. Typically, the video decoder include only standard specification, technical research mode selection falls under the category of the encoder, it is not listed in the standards.
Be
AVC / H.264 Profile AVC / H.264 defines four different Profile (class): Baseline (Baseline), Main (main class), Extended (extended class) and High Profile (high profile) (under their respective into a plurality of layers): - Baseline Profile provides I / P frame, only supports progressive (progressive scan) and CAVLC - Extended Profile provides I / P / B / SP / SI-frame, only supports progressive (progressive scan) and CAVLC - Main Profile provides I / P / B frames support progressive (progressive scan) and interlaced (interlaced), provided as CAVLC or CABAC - High Profile (i.e. FRExt) added on the basis of the Main Profile: 8x8 intra prediction (8x8 intra prediction), custom quant (custom quantization), lossless video coding (lossless video coding), more yuv format (4: 4: 4 ...) below the site to introduce very detailed, http: // www .tfcc.cn / dvdrip / 2006-6-15_14-13-02_2528.html
================================================== ==============================================
MPEG-4 AVC / H.264 information
Transfer from http://www.tfcc.cn/dvdrip/2006-6-15_14-13-02_2528.html
This series consists of FAQ (haibara) translation, by kaito_mkid (pcicp) help during to thank, because of Newbie, if the translation error, please also pointed out that,
Compiled by Doom9's original bond.
Translator's Note: The basic professional name I have to be interpreted in parentheses in the form of detailed information please refer to the information on their own.
What is MPEG-4? MPEG-4 (ISO 14496) developed by our well-known MPEG-1 (MP3, VCD) and MPEG-2 (DVD, SVCD) standard, also argued for the various audio / video compression formats and more standardized international organization for standardization (international organization for standardization / ISO) working group, namely Moving picture Experts group (Moving picture Experts group / MPEG) development of a wide range of open standards.
MPEG-4 standard is essentially the only target of standardization some promising products (eg DVD achieve performance), but also covers a wide range of sub-standard product allows a vendor to select the appropriate standard required to perform.
MPEG-4 standard, as noted above, is divided into many different sub-criteria, the following partial Doom9 where we are most interested users:
- ISO 14496-1 (Systems / system)
Animation / interactivity (like DVD menu)
- ISO 14496-2 (Video / Video)
Such as Advanced Simple Profile (ASP / Advanced Simple Profile) from XviD, DivX5, 3ivx ... execution
- ISO 14496-3 (Audio / audio) , Advanced Audio Coding (AAC / Advanced Audio Coding)
- ISO 14496-10 (Video / Video) , Advanced Video Coding (AVC / Advanced Video Coding), there are well-known H.264
- ISO 14496-14 (Container / vessel) , MP4 container format (for use .mp4 extension)
- ISO 14496-17 (Subtitles / Subtitled) , MPEG-4 format text subtitle timing
These information pages a positive force in most of the useful information provided these criteria, especially as the focus of MPEG-4 ASP and AVC / H.264.
Compared with closed formats such as Micro $ oft's Windows Media, MPEG-4 as the same open standards for what possible advantage?
One of the benefits of open standards that allow when creating products
Open to all
And follow it. So we have many different products in line with MPEG-4 standard and compatible with each other.
Followed by
interoperability
and
A wide range of products
Selectivity, leading to an open standard
compete , Which means that the customer in terms of products in a competitive market can better improvequality ,less price And more focus on the customer need. But do not forget this for us, perhaps the most important point:
An open standard allows
Open Source Development , For example, we know XviD.
ISO 14496-10 (Video) - Advanced Video Coding (Advanced Video Coding / AVC)
MPEG-4 standard with AVC / H.264 defines a have the latest, highest technology and practical, full of cutting-edge technology of video encoding format.
In 2003, ISO's MPEG (Moving Pictures Experts Group / Moving Picture Experts Group) and the UN (United Nations / UNO) under the ITU (International Telecommunication Union / ITU) of VCEG (Video Coding Experts Group / Video Coding Experts Group ) the two major tissue finalized and the predetermined co-AVC / H.264 video coding standard, wherein the latter has a standardized format H.263 (videoconferencing now for the primary). Be
Joint Video Team AVC / H.264 standard itself is made up of experts MPEG and VCEG (JVT / Joint Video Team) development. Be Run by the MPEG, the standard is called MPEG-4 Part 10 (ISO 14496-10); run by the ITU, it is called the H.264 (ITU number of documents) wherein as the latter has been widely spread. The new criteria for selecting Advanced Video Coding (AVC / Advanced Video Coding) as the "official" name - because the corresponding video audio format is Advanced Audio Coding (AAC / Advanced Audio Coding).
AVC / H.264 Profil AVC / H.264 defines four different Profile (class): Baseline (Baseline), Main (major class), Extended (extended type) and High Profile (high class) (each of which is divided into a plurality of next Floor - Baseline Profile Providing I / P frame, only supports Progressive (progressive scan) and CAVLC Extended Profile Providing I / P / B / SP / SI-frame, only supports Progressive (progressive scan) and CAVL - Main Profile Providing I / P / B frames support Progressive (progressive scan) and interlaced (interlaced), providing CAVLC or CABAC - High Profile(Ie FRExt) on the basis of the new Main Profile: 8x8 intra prediction (8x8 intra prediction), custom quant (custom quantization), lossless video coding (lossless video coding), more yuv format (4: 4: 4 ...)Only in the future will tell us which frameworks and tools are most likely to be used for DVD backups, but I guess will have the following tools to help main and / or high profile (also please verify
MPEG-4 ASP
Tool description, in addition to GMC, all tools are available under AVC):
CAVLC / CABAC:
AVC / H.264 defines two kinds of advanced reconstruction data streams (including Macroblock-Type, MotionVectors + Reference-Index (Reference Index) ...) entropy encoding tools, beyond MPEG -4 ASP: Context-Adaptive Variable Length Coding (CAVLC / Context Adaptive Biometric Coding) and CONTEXT-Adaptive Binary Arithmetic Coding (Cabac / Context Adaptive Binary Arithmetic Coding) Cabac, just the default algorithm in AVC / H.264 Compared to CAVLC (also called UVLC), it is a more powerful algorithm, which is said to reduce the bit rate of 10-15% (especially at high yard ratio). Cabac (like CAVLC) is a non-destructive algorithm, so it does not reduce the quality, but slows the encoding and decoding speed loop / deblocking filter (loop / deck filter): with the pretreatment filter (for example by Avisynth End), or after processing filter (by decoder acting on the final output), the circulator filter is filtered over the encoding process, and the reference is used for the next after the encoding is completed. It can eliminate block effects, especially at low yards, but also slow down the encoding and decoding speed. Variable Block Sizes / MacRoblock Partitions (block size): AVC / H.264 dynamic search accuracy compared to MPEG-4 ASP (INTER4V / 4MV can be 16X16 and 8x8 block transform) (Motion Search Precision) Raise to 4x4 (including 8x4 ...). The size of the sub-block is adaptive / variable, and an excellent encoder should be able to make the most appropriate division size for each macroblock. Multiple Reference Frames: Compared to MPEG-4 ASP (previously used as a reference to reference), AVC / H.264 uses inter-frame search mode and provides a reference to the reference. It means that the AVC / H.264 encoding can decide whether it is just simply referring to before or even the reference is more than before. Therefore, (for example, a certain P 桢 can refer to a nearest I before I have to propose a new type of type: IDR, it is the subsequent 不 桢 桢. Using multiple references will slow down coding and decoding speed and may result in only at the IDR clip. Weighted prediction: With weighted predictions to weighted reference (for example, you can measure the brightness of the previous image). It is particularly helpful in the pre-image of the image that may be fade in or fade out at any time, just the processing of the image dimmed. WP does not have effects on the image that appears at the same time (for example, when the image turns from a scene to another, it will fade in simultaneously).
Rate Distormion Optimisation (RDO / Code Rate - Distortion / Loss Optimization): As long as the encoder has to make an extractive in numerous selection, RDO allows the most effective coding method (for example, a dynamic search in the interruption / Decision ...) RDO is not the tool defined by the AVC / H.264 itself. It is the first tool introduced by the H.264 reference software, which is a new way. Other encodings can also utilize RDOs, such as XVID 'VHQ modes, using RDO AVC / H.264 and other popular video encoding formats: existing AVC / H.264 encoding can be used for end users AVC / H. 264 Executors have X264, Nero, Apple, Sorenson, Elecard, Moonlight, VSS, MPEGABLE, Envivio, HDOT264
(Binary), DSPR, JM (reference software), FFmpeg, Philips, Fastvdo, SKAL, SONY, and more encoder x264: First publicly available encoder, open source (GPL) (source code) ), Using VFW: X264VFW, FfdShow, command line: x264Cli (output RAW. MP4), Mencoder (output RAW .avi) (MEGUI) or FFMPEG X264 support 2Pass, Cabac, Circulating filter, multiple B frame, bidewire reference (B-Reference), multi-reference, 4x4 P frame block transformation, 8x8 B frame block division, distortion signaling (ANAMORPHIC SIGNALLING) and High Profile: 8x8 DTC (discrete cosine transformation) Inframe prediction, lossless video coding and custom lossless quantization matrix Nerodigital AVC: Nero Recoded is included, output is. MP4 ND AVC support 2Pass, Cabac, adaptive cycle filter, multiple B frame, multiple references, Weighted prediction, 8x8 B frame division, 16x16 B block division, adaptive quantization (height psy) Sorenson: Sorenson SQUEEZE 4 contains, output .mp4, sorenson support 2Pass, maximum 2 consecutive B frame, two-way reference, cycling filtering and Multiple SLICE Apple: QuickTime 7 Contains, output .mp4, .3gp and .mov, 2Pass, maximum 1 consecutive B frame, cyclic filter (0, 0), 8x8 P frame block transform, 8x8 B block block Transform, 4x4 I Frame Transformation, Adaptive Quantization, Level 5 SLICE, No CABAC, No Weighted Prediction, No Multiple References JM: This AVC Reference Software V9.3 provides Main and High Profile: B / SP frame, Cabac, Circulating filter, 4x4 block division, multi-reference, adaptive quantization, error repair (ERROR Resilience), RDO, lossless video coding, custom quantization, ratio control (Rate Control), etc. ... HDOM9 Member CharacT3R Development Source (GPL) VFW version of the reference software, still with very old reference soft Part (JM 4.0C) is the foundation VSS: Free VFW encoder (5 days preview), the reference encoder is based on the Elecard: electrow converter, output .mp4, mainconcept's v2 encoder, Output .264 And. MPG PS / TS is no longer public: Moonlight: Moonlight's OneClick Compressor V1. And Cyberlink's Powerencoder, output. MPG Moonlight supports 1Pass (VBR / CBR / preset quantization), Cabac, cyclic filtering, maximum 2 Continuous B frame, 8x8 B block division, adaptive quantization, automatic adjustment PAR (Pixel Aspect Ratio / Pixel Aspect ratio), interlaced Mainconcept: Free encoder (with watermark), output .264 and .mpg PS / TS
1Pass (CBR / VBR / preset quantization), p-frame reorder, CABAC, cycle filtering, multiple B-VOPs (Bidirectional Video Object Planes / two-way video object plane), multiple reference, 4x4 P frame Block division), PAR, RDO MPEGABLE: Provides a limited free VFW encoder (not based on the reference software), does not support YV12 MPEGABLE support only 1Pass (preset quantization), 8x8 p frame block, only Support CAVLC, cyclic filter Envivio: 4cod, output .mp4 decoder (comparative evaluation) FFmpeg: Open Source (LGPL), included in FFDSHOW (VFW and DSHOW Encoder), MPLAYER and VIDEOLAN, etc. Support B, two-way reference, CABAC , Cycle filtering, weighting prediction, and high profile (8x8 DCT and intra prediction, lossless video coding) Apple: QuickTime 7 supports AVC decoding, support. MP4 / .MOV, decoding speed is slow only to support up to 1 Continuous B frame, Cabac, Recycling, multiple B frames, multiple B frames, and interlaced, NERODIGITAL AVC: Recode2 come with DSHOW decoder and .mp4 Parser support Main and High Profile VSS: VFW decoder (preview 5 days) and DSHOW decoder (limited 30 days) VSS DSHOW Support. AVI (support VSSH and H264 Fourcc), Cabac, cyclic filtering, B frame electrodard: electroncept's v2 encoder included ENVIVIO: A non-free AVC DSHOW decoder named EnviviotV, From version 2.0, the current version is 2-1-181) Support Mix AVC. MP4 Philips: AVC Alliance Player with free DSHOW AVC decoder (RAW AVC) Fastvdo: Time Limit (Each video file 5 minutes) High profile dshow decoder PegaSus: Here is an immature DSHOW AVC decoder Basic AVC Decoder to write with C, a university engineering, see here
Our other product:
