CN116828198B - Method for supporting VA-API hardware video acceleration interface on NVIDIA GPU - Google Patents
Method for supporting VA-API hardware video acceleration interface on NVIDIA GPU Download PDFInfo
- Publication number
- CN116828198B CN116828198B CN202311096314.1A CN202311096314A CN116828198B CN 116828198 B CN116828198 B CN 116828198B CN 202311096314 A CN202311096314 A CN 202311096314A CN 116828198 B CN116828198 B CN 116828198B
- Authority
- CN
- China
- Prior art keywords
- video
- interface
- api
- request
- nvenc
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000001133 acceleration Effects 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 14
- 230000006870 function Effects 0.000 claims description 51
- 238000013507 mapping Methods 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 abstract description 4
- 238000012545 processing Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000005314 correlation function Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/42—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/44—Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The method for supporting the VA-API hardware video acceleration interface on the NVIDIA GPU comprises the steps of constructing a compatible layer between the VA-API interface and the NVDEC/NVENC interface, analyzing the request of the VA-API interface, calling the NVDEC/NVENC interface related function based on the analyzed request to complete the related request and returning the related request to the VA-API interface; in video decoding, a request that a compatible layer can parse involves the steps of: querying a supported video coding specification, creating a video coder-decoder, decoding video frames, reading decoded frame data and displaying, and destroying the video decoder. By means of the hardware video acceleration interface compatible layer, the invention provides the VA-API hardware video acceleration interface for the video program, realizes the conversion from the VA-API hardware video acceleration interface to the NVDEC/NVENC under Linux, realizes the hardware video acceleration function by means of the NVIDIA special drive and the GPU, reduces the occupation of the video program to CPU resources, and improves the user experience.
Description
Technical Field
The invention relates to the technical field of hardware video function compatibility, in particular to a method for supporting a VA-API hardware video acceleration interface on an NVIDIA GPU.
Background
Hardware video acceleration refers to encoding and decoding video by using a GPU, so that the load of a CPU is reduced, and the user experience is improved. To save space and bandwidth, video is usually compression encoded in a specific format before storage and transmission, whereas reverse decoding is also required during playback. The encoding and decoding of video involves a large amount of operations, and if software encoding and decoding are used, a large amount of CPU resources are consumed, which may affect the normal operation of the system. In modern computers, a video encoding and decoding module is generally built in a GPU, if the GPU encoding and decoding module is used, GPU resources can be fully utilized, CPU computing power is released, and the resource utilization rate of a system is improved.
Currently, on the Linux operating system, popular and general hardware video acceleration interfaces mainly comprise VA-API, VDPAU and the like. VA-API is proposed by Intel corporation and has become the mainstream scheme under Linux at present; the VDPAU was proposed by NVIDIA, and after NVIDIA pushed out a new proprietary interface NVDEC/NVENC, the development of the project has been nearly stalled.
NVIDIA GPU is the most popular GPU in the current market, and the hardware video acceleration interface under Linux is mainly NVDEC/NVENC interface. Since the interface is a NVIDIA dedicated interface, many Linux programs are not supported (such as Firefox, etc.), and thus, the hardware video acceleration function cannot be realized by using the NVIDIA GPU.
Chinese patent No. CN109963191A, a method and a device for processing video information. The embodiment of the invention discloses a method, a device and a storage medium for processing video information, wherein the method comprises the following steps: acquiring video stream data, and determining a first hard decoding API encapsulated by adopting a first interface specification; and calling the first hard decoding API to obtain a second hard decoding API packaged by adopting a second interface specification, wherein the second hard decoding API is contained in a hard decoding library corresponding to the first hard decoding API and is used for calling the hard decoding library in the display card drive to decode the video stream data. By the scheme of the embodiment of the invention, the CPU occupancy rate of the terminal equipment can be effectively reduced, and the online video playing performance of the terminal equipment is improved. However, this invention also has significant disadvantages: the interface adaptation library realized by the invention converts the VAPAU interface into a VA-API interface. Considering that in the scene that the Linux operating system is matched with the NVIDIA GPU, the application program is mainly supported by VA-API, and the interface provided by the NVIDIA GPU is mainly NVDEC/NVENC, the method is not applicable any more, and the hardware video acceleration function cannot be provided for the application program.
Chinese invention patent "video frame processing method and apparatus" (patent number: CN 114245138A). The invention provides a video frame processing method and a video frame processing device. In the method, transcoded data for the video frame is obtained. The transcoded data is stored in a designated address space in the GPU. Thereafter, a first mapping relationship is established between a pointer to the specified address space and an index number of a target texture to be created in the GPU. Then, the transcoding data stored in the designated address space is created into a target texture according to the first mapping relation. However, this invention also has significant disadvantages: in the method of the invention, an NVDEC/NVENC interface is adopted, vA-API is a mainstream hardware video acceleration interface in Linux, most video programs support VA-API instead of NVDEC/NVENC, and therefore, the application scene of the invention is also greatly limited.
At present, the hardware video acceleration interface of the Linux operating system is mainly VA-API, and the hardware video acceleration interface provided by the NVIDIA GPU special driver is mainly NVENC/NVDEC. This means that many video related applications on Linux (e.g., firefox, etc.) cannot achieve hardware video acceleration with NVIDIA GPU. The existence of the problem causes that the encoding and decoding capability of the GPU cannot be utilized, and a large amount of CPU resources are consumed, so that the normal operation of the system is affected.
In summary, it is needed to invent a video hardware acceleration device, so that a video program based on VA-API on Linux can realize a hardware video acceleration function through an NVDEC/NVENC interface by means of an NVIDIA GPU.
Disclosure of Invention
In order to solve the defects existing in the prior art, the invention provides a method for supporting a VA-API hardware video acceleration interface on an NVIDIA GPU, which is characterized in that a compatible layer is constructed between a VA-API interface and an NVDEC/NVENC interface and is used for analyzing a request of the VA-API interface, calling an NVDEC/NVENC interface correlation function based on the analyzed request to complete the correlation request and returning the correlation request to the VA-API interface;
in video decoding, hardware video acceleration is achieved by:
step S1: query supported video coding specification: the video program inquires the supported coding specification from the compatible layer, and the compatible layer analyzes the request and inquires about the relevant characteristics; returning the query result and transmitting the query result to the video program in an original way;
step S2: creating a codec, wherein the video program initiates a request for creating a video codec context to a compatible layer, and the compatible layer parses the request and creates a decoder in a corresponding format;
step S3: decoding the video frame, the video program initiating a request for decoding the video frame to a compatible layer, the compatible layer parsing the request and decoding the video frame;
step S4: the decoded frame data is read and displayed, the compatible layer copies the frame data into the VA-API buffer, and the video program displays the decoded frame data in the VA-API buffer to the program window.
In the step S1, the video program inquires the supported coding specification from the compatible layer by calling the related function of the VA-API interface;
in the step S2, the video program initiates a request for creating a video codec context to a compatible layer by calling a VA-API interface related function;
in the step S3, the video program initiates a video frame decoding request to a compatible layer by calling a VA-API interface related function;
in step S4, the video program displays the decoded frame data in the VA-API buffer to the program window by calling the VA-API interface correlation function.
In the step S1, the compatible layer sequentially queries the NVIDIA GPU for relevant characteristics by calling the NVDEC/NVENC interface relevant functions to NVIDIA Video Codec SDK and NVIDIA Linux Driver;
in the step S2, the compatible layer creates a decoder with a corresponding format by calling the related function of the NVDEC/NVENC interface;
in the step S3, the compatible layer decodes the video frame by calling the related function of the NVDEC/NVENC interface;
in step S4, the compatible layer copies the frame data into the VA-API buffer by calling NVDEC/NVENC interface-related functions.
Wherein, still include the following step:
step S5: the video program judges whether all video frames are decoded, if not, the video program jumps to the step S3 to continue decoding, if so, the next step is entered;
step S6: the video program calls the VA-API interface related function to initiate a destroy video decoder request to the compatible layer, which parses the request and calls the NVDEC/NVENC interface related function to destroy the video decoder.
The invention provides a VA-API hardware video acceleration interface for video programs by means of a hardware video acceleration interface compatible layer, receives and analyzes video requests of the video programs, and then calls an NVDEC/NVENC interface to complete corresponding video encoding and decoding operations by means of an NVIDIA GPU. The conversion from the hardware video acceleration interface VA-API to the NVDEC/NVENC under the Linux is realized, so that the video program based on VA-API can call the NVDEC/NVENC interface, the hardware video acceleration function is realized by means of the NVIDIA special drive and the GPU, the occupation of the video program to CPU resources is reduced, and the user experience is improved.
Detailed Description
In order to further understand the technical scheme and beneficial effects of the present invention, the technical scheme and beneficial effects thereof will be described in detail below.
The terms involved in the present invention are explained as follows:
GPU: graphics processor, collectively Graphics Processing Unit. The display core, the vision processor and the display chip are also called as a microprocessor which is specially used for performing image and graph related operation on personal computers, workstations, game machines and some mobile devices (such as tablet computers, smart phones and the like).
VA-API: a general hardware video acceleration interface, collectively Video Acceleration API, is proposed by Intel corporation.
libva: a hardware video acceleration library is an open source implementation of a VA-API hardware video acceleration interface, and is developed by Intel corporation.
VADPU: a general hardware video acceleration interface, collectively Video Decode and Presentation API for Unix, is proposed by NVIDIA corporation, and is currently lacking in maintenance and has been developed with little stagnation.
NVDEC/NVENC: a special hardware video acceleration interface is proposed by NVIDIA company and is the interface specification currently adopted by NVIDIA company.
NVIDIA Video Codec SDK: a video codec development suite special for NVIDIA company provides NVDEC/NVENC interface.
NVIDIA Linux Driver: NVIDIA provides a special GPU driver for the Linux operating system.
Nv_code_api_function_list: NVIDIA Video Codec SDK is a set of code interface function tables, which consists of a plurality of code related interface functions.
Aiming at the difficult problems existing in the prior art, the invention creates a new hardware video acceleration interface compatible layer (hereinafter referred to as compatible layer) which runs between the NVDEC/NVENC interface and the VA-API interface of the video program, realizes the interface conversion from VA-API to NVDEC/NVENC under Linux, ensures that the video program based on VA-API can call the NVDEC/NVENC interface, realizes hardware video acceleration by means of NVIDIA special drive and GPU, reduces the occupation of CPU resources by the video program and improves the user experience.
Taking video decoding as an example, the specific implementation steps of the invention are as follows:
1. the supported video coding specification is queried. The video program calls VA-API interface functions vaQueryConfigProfiles and vaQueryConfigEntry points to inquire the supported coding specifications from a compatible layer; after receiving the query request, the compatible layer analyzes the request and calls the NVDEC/NVENC interface function cuvidGetDecoderCap to query NVIDIA Video Codec SDK for the supported coding specification; NVIDIA Video Codec SDK continues to call NVIDIA Linux Driver the correlation interface to query the NVIDIA GPU for correlation properties; the GPU returns the query result and transmits the query result to the video program in the original path; and the video program judges whether the system supports the corresponding format according to the returned result, if so, the video program enters the next step, and if not, the video program exits.
2. A codec is created. The video program calls VA-API interface functions vaCreateConfig and vaCreateContext to initiate a create video codec context request to the compatible layer; the compatibility layer parses the request and calls the NVDEC/NVENC interface function cuvidcatedecoder to create a decoder of the corresponding format.
3. The video frames are decoded. The video program calls the VA-API interface function vaBeginPicture, vaRenderPicture and vanadpicture to initiate a request to the compatibility layer to decode a video frame; the compatibility layer parses the request and invokes the NVDEC/NVENC interface function cuviddecodefpicture to decode the video frame.
4. The decoded frame data is read and displayed. The compatible layer calls the NVDEC/NVENC interface function cuvidMapvideo frame to carry out memory mapping on a buffer area storing decoded frame data, then copies the frame data into a corresponding VA-API buffer area, and finally calls the cuvidUnmapvideo frame to remove the memory mapping; the video program calls the VA-API interface function vaPutSurface to display the decoded frame data in the VA-API buffer to the program window.
5. The video program judges whether all video frames are decoded, if not, the video program jumps to the step 3 to continue decoding, and if so, the video program goes to the next step.
6. Destroying the video decoder. The video program calls VA-API interface functions vardiststroyContext and vardiststroyConfig to initiate a request to the compatible layer to destroy the video decoder; the compatibility layer parses the request and invokes the NVDEC/NVENC interface function cuvidDestroydecoder to destroy the video decoder.
7. And (5) finishing decoding.
Taking video coding as an example, the specific steps are as follows:
1. and loading the code interface function table. The compatibility layer calls NVDEC/NVENC interface FUNCTION nvencoAN_SNetigmaxsupplortedversion to query NVIDIA Video Codec SDK for the version of the encoding interface and calls nvencoAN_apicreatelist interface load encoding interface FUNCTION table nv_code_api_function_list.
2. The supported coding specification is queried. The video program calls VA-API interface functions vaQueryConfigProfiles and vaQueryConfigEntry points to inquire the supported coding specifications from a compatible layer; after receiving the inquiry request, the compatible layer analyzes the request and calls NVDEC/NVENC interface FUNCTION NV_ENCODE_API_FUNCTION_LIST that nvEncGetEncodeeCaps inquires NVIDIA Video Codec SDK about the supported coding specification; NVIDIA Video Codec SDK continues to call NVIDIA Linux Driver the correlation interface to query the NVIDIA GPU for correlation properties; the GPU returns the query result and transmits the query result to the video program in the original path; and the video program judges whether the system supports the corresponding format according to the returned result, if so, the video program enters the next step, and if not, the video program exits.
3. A video encoder is created. The video program calls VA-API interface functions vaCreateConfig and vaCreateContext to initiate a create video codec context request to the compatible layer; the compatibility layer parses the request and calls NVDEC/NVENC interface FUNCTION nv_code_api_function_list:: nvencopenencoderless ex creates an encoder of the corresponding format and calls nv_code_api_function_list::: NVENC inititizer encoder initialization encoder.
4. Video frames are encoded. The video program calls the VA-API interface function vaBeginPicture, vaRenderPicture and the vanadpicture to initiate a request to the compatibility layer to encode a video frame; the compatible layer parses the request and calls NVDEC/NVENC interface FUNCTION NV_ENCODE_API_FUNCTION_LIST, which is nvEncEncEncodieTight encoding video frame; NVIDIA Video Codec SDK encodes the video frames and stores the encoded bitstream data into a designated bitstream buffer after completion.
5. The encoded bitstream data is read. The compatible layer calls NVDEC/NVENC interface FUNCTION NV_ENCODE_API_FUNCTION_LIST that nvEncLockBitstream locks the bit stream buffer, copies the encoded bit stream data from the bit stream buffer into the corresponding VA-API buffer, and finally calls NV_ENCODE_API_FUNCTION_LIST that nvEncLockBitstream unlocks the bit stream buffer; the video program calls the VaMapBufer mapping VA-API buffer area of the VA-API interface function, reads the coded bit stream data in the buffer area, stores the data into a memory or carries out network transmission according to the user demand, and finally calls the VaUnmapBuffer to remove the buffer area mapping.
6. And the video program judges whether all video frames are coded, if not, the video program jumps to the step 4 to continue coding, and if so, the video program goes to the next step.
7. Destroying the video encoder. The video program calls VA-API interface functions varDestroyContext and varDestroyConfig to initiate a request for destroying the video encoder to the compatible layer; the compatibility layer parses the request and calls the NVDEC/NVENC interface FUNCTION NV_ENCODE_API_FUNCTION_LIST that nvEncDestroyEncoder destroys the video encoder.
8. The encoding is ended.
The invention provides a VA-API hardware video acceleration interface for video programs by means of a hardware video acceleration interface compatible layer, receives and analyzes video requests of the video programs, and then calls an NVDEC/NVENC interface to complete corresponding video encoding and decoding operations by means of an NVIDIA GPU. The conversion from the hardware video acceleration interface VA-API to the NVDEC/NVENC under the Linux is realized, so that the video program based on VA-API can call the NVDEC/NVENC interface, the hardware video acceleration function is realized by means of the NVIDIA special drive and the GPU, the occupation of the video program to CPU resources is reduced, and the user experience is improved.
The present invention supports both video decoding and encoding operations. And multiple codec specifications may be supported simultaneously, including, but not limited to, h.264, h.265, MPEG-2, VC1, VP9, etc., with specific support being determined by NVIDIA GPU and NVIDIA Video Codec SDK.
Although the present invention has been described with reference to the above preferred embodiments, it should be understood that the present invention is not limited to the above embodiments, and that various changes and modifications can be made herein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (2)
1. The method for supporting the VA-API hardware video acceleration interface on the NVIDIA GPU is characterized by comprising the following steps of: constructing a compatible layer between the VA-API interface and the NVDEC/NVENC interface, analyzing the request of the VA-API interface, calling the NVDEC/NVENC interface related function to complete the related request based on the analyzed request, and returning to the VA-API interface;
in video decoding, hardware video acceleration is achieved by:
step S1: query supported video coding specification: the video program calls the VA-API interface related function to inquire the supported coding specification from the compatible layer, the compatible layer analyzes the request and calls the NVDEC/NVENC interface related function to inquire the related characteristics from NVIDIA Video Codec SDK and NVIDIA Linux Driver to the NVIDIA GPU in sequence; the GPU returns the query result and transmits the query result to the video program in the original path;
step S2: creating a codec, wherein the video program calls a VA-API interface related function to initiate a request for creating a video codec context to a compatible layer, and the compatible layer analyzes the request and calls an NVDEC/NVENC interface related function to create a decoder in a corresponding format;
step S3: decoding the video frame, the video program calls the VA-API interface related function to initiate a request for decoding the video frame to a compatible layer, and the compatible layer analyzes the request and calls the NVDEC/NVENC interface related function to decode the video frame;
step S4: and reading and displaying the decoded frame data, calling an NVDEC/NVENC interface related function by a compatible layer to perform memory mapping on a buffer for storing the decoded frame data, copying the frame data into a VA-API buffer, finally removing the memory mapping, and calling the VA-API interface related function by a video program to display the decoded frame data in the VA-API buffer to a program window.
2. The method of supporting a VA-API hardware video acceleration interface on an NVIDIA GPU of claim 1, wherein: the method also comprises the following steps:
step S5: the video program judges whether all video frames are decoded, if not, the video program jumps to the step S3 to continue decoding, if so, the next step is entered;
step S6: the video program calls the VA-API interface related function to initiate a destroy video decoder request to the compatible layer, which parses the request and calls the NVDEC/NVENC interface related function to destroy the video decoder.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311096314.1A CN116828198B (en) | 2023-08-29 | 2023-08-29 | Method for supporting VA-API hardware video acceleration interface on NVIDIA GPU |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311096314.1A CN116828198B (en) | 2023-08-29 | 2023-08-29 | Method for supporting VA-API hardware video acceleration interface on NVIDIA GPU |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116828198A CN116828198A (en) | 2023-09-29 |
| CN116828198B true CN116828198B (en) | 2023-11-28 |
Family
ID=88122500
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311096314.1A Active CN116828198B (en) | 2023-08-29 | 2023-08-29 | Method for supporting VA-API hardware video acceleration interface on NVIDIA GPU |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116828198B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109963191A (en) * | 2017-12-14 | 2019-07-02 | 中兴通讯股份有限公司 | A kind of processing method of video information, device and storage medium |
| CN110471766A (en) * | 2019-08-06 | 2019-11-19 | 北京华恒盛世科技有限公司 | A kind of GPU resource scheduling system and method based on CUDA |
| CN112650984A (en) * | 2019-10-10 | 2021-04-13 | 百度（美国）有限责任公司 | Data processing accelerator and computer-implemented method executed by the same |
| CN113852751A (en) * | 2020-06-28 | 2021-12-28 | 腾讯科技（深圳）有限公司 | Image processing method, device, terminal and storage medium |
| CN116097224A (en) * | 2021-04-15 | 2023-05-09 | 辉达公司 | Simultaneous boot code |
| CN116400998A (en) * | 2022-09-08 | 2023-07-07 | 麒麟软件有限公司 | Video hardware acceleration device and method suitable for virtual display card |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100253578A1 (en) * | 2007-11-25 | 2010-10-07 | Mantovani Jose R B | Navigation data acquisition and signal post-processing |
| US8539039B2 (en) * | 2010-06-22 | 2013-09-17 | Splashtop Inc. | Remote server environment |
| US20230185637A1 (en) * | 2021-12-10 | 2023-06-15 | Nvidia Corporation | Application programming interfaces for interoperability |
-
2023
- 2023-08-29 CN CN202311096314.1A patent/CN116828198B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109963191A (en) * | 2017-12-14 | 2019-07-02 | 中兴通讯股份有限公司 | A kind of processing method of video information, device and storage medium |
| CN110471766A (en) * | 2019-08-06 | 2019-11-19 | 北京华恒盛世科技有限公司 | A kind of GPU resource scheduling system and method based on CUDA |
| CN112650984A (en) * | 2019-10-10 | 2021-04-13 | 百度（美国）有限责任公司 | Data processing accelerator and computer-implemented method executed by the same |
| CN113852751A (en) * | 2020-06-28 | 2021-12-28 | 腾讯科技（深圳）有限公司 | Image processing method, device, terminal and storage medium |
| CN116097224A (en) * | 2021-04-15 | 2023-05-09 | 辉达公司 | Simultaneous boot code |
| CN116400998A (en) * | 2022-09-08 | 2023-07-07 | 麒麟软件有限公司 | Video hardware acceleration device and method suitable for virtual display card |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116828198A (en) | 2023-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI393065B (en) | Method and system of operating frame buffer manager, decoding and displaying compressed video data | |
| US7447266B2 (en) | Decoding device and decoding program for video image data | |
| CN111885346B (en) | Picture code stream synthesis method, terminal, electronic device and storage medium | |
| CN113055681B (en) | Video decoding display method and device, electronic equipment and storage medium | |
| US11483561B2 (en) | Transform method in picture block encoding, inverse transform method in picture block decoding, and apparatus | |
| CN111741232A (en) | Method for improving ultra-high-definition non-editing performance based on dual-display card NVLINK | |
| CN110933495A (en) | Video playing method and device based on embedded system | |
| US20240098310A1 (en) | Encoding method, real-time communication method, apparatus, device, and storage medium | |
| CN112714320A (en) | Decoding method, decoding device and computer readable storage medium | |
| CN116400998A (en) | Video hardware acceleration device and method suitable for virtual display card | |
| CN116828198B (en) | Method for supporting VA-API hardware video acceleration interface on NVIDIA GPU | |
| CN116450149B (en) | Hardware decoding method, device and storage medium | |
| KR20170069074A (en) | An operation method of a decoder, and an operation method of an application processor including the decoder | |
| CN103037209A (en) | Method and device for decoding and processing video frame | |
| US5883679A (en) | Scanning scheme for images stored in dynamic random access memory | |
| CN117130749A (en) | Method for improving hardware decoding capability of Web player based on WebGPU | |
| CN114173154B (en) | Video processing method and system | |
| US20140096168A1 (en) | Media Playing Tool with a Multiple Media Playing Model | |
| CN115460183A (en) | Method and device for playing streaming media data based on browser and related equipment | |
| CN111556365A (en) | Video transcoding method and device | |
| CN114584786B (en) | Memory allocation method and system based on video decoding | |
| CN117880593A (en) | Network video stream real-time display system and display method based on RK3588 processing chip | |
| US20240022743A1 (en) | Decoding a video stream on a client device | |
| CN101583035B (en) | Access method, device and system of audio frequency and video file | |
| CN106341691A (en) | Big data-based data processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |