Represents one raw (undecoded) picture in a video stream, plus a timestamp for when to display that video picture relative to other items in a IContainer. More...

#include <IVideoPicture.h>

Inheritance diagram for com::avpkit::core::IVideoPicture:

[legend]

Collaboration diagram for com::avpkit::core::IVideoPicture:

[legend]

Public Types
enum	PictType { DEFAULT_TYPE =0 , I_TYPE = 1 , P_TYPE = 2 , B_TYPE = 3 , S_TYPE = 4 , SI_TYPE = 5 , SP_TYPE = 6 , BI_TYPE = 7 }
	The different types of images that we can set. More...

enum	FrameDataType { AV_FRAME_DATA_PANSCAN , AV_FRAME_DATA_A53_CC , AV_FRAME_DATA_STEREO3D , AV_FRAME_DATA_MATRIXENCODING , AV_FRAME_DATA_DOWNMIX_INFO , AV_FRAME_DATA_REPLAYGAIN , AV_FRAME_DATA_DISPLAYMATRIX , AV_FRAME_DATA_AFD , AV_FRAME_DATA_MOTION_VECTORS , AV_FRAME_DATA_SKIP_SAMPLES , AV_FRAME_DATA_AUDIO_SERVICE_TYPE , AV_FRAME_DATA_MASTERING_DISPLAY_METADATA , AV_FRAME_DATA_GOP_TIMECODE , AV_FRAME_DATA_SPHERICAL , AV_FRAME_DATA_CONTENT_LIGHT_LEVEL , AV_FRAME_DATA_ICC_PROFILE }

Public Member Functions
virtual bool	isKeyFrame ()=0
	Is this a key frame? More...

virtual void	setKeyFrame (bool aIsKey)=0
	Reset if this is a key frame or not. More...

virtual bool	isComplete ()=0
	Is this picture completely decoded? More...

virtual int32_t	getSize ()=0
	Total size in bytes of the decoded picture. More...

virtual int	getWidth ()=0
	What is the width of the picture. More...

virtual int	getHeight ()=0
	What is the height of the picture. More...

virtual IPixelFormat::Type	getPixelType ()=0
	Returns the pixel format of the picture. More...

virtual int64_t	getPts ()=0
	What is the Presentation Time Stamp (in Microseconds) of this picture. More...

virtual void	setPts (int64_t value)=0
	Set the Presentation Time Stamp (in Microseconds) for this picture. More...

virtual int	getQuality ()=0
	This value is the quality setting this VideoPicture had when it was decoded, or is the value to use when this picture is next encoded (if reset with setQuality() More...

virtual void	setQuality (int newQuality)=0
	Set the Quality to a new value. More...

virtual int	getDataLineSize (int lineNo)=0
	Return the size of each line in the VideoPicture data. More...

virtual void	setComplete (bool aIsComplete, IPixelFormat::Type format, int width, int height, int64_t pts)=0
	After modifying the raw data in this buffer, call this function to let the object know it is now complete. More...

virtual bool	copy (IVideoPicture *srcPicture)=0
	Copy the contents of the given picture into this picture. More...

virtual void	render (bool drop=false, int64_t timeStamp=-1)=0
	Render this picture on configured surface. More...

virtual void *	getOpaqueData ()=0

virtual PictType	getPictureType ()=0
	Get the picture type. More...

virtual void	setPictureType (PictType type)=0
	Set the picture type. More...

virtual void	setSideData (FrameDataType type, com::avpkit::ferry::IBuffer *buffer)=0

Public Member Functions inherited from com::avpkit::core::IMediaData
virtual int64_t	getTimeStamp ()=0
	Get the time stamp of this object in getTimeBase() units. More...

virtual void	setTimeStamp (int64_t aTimeStamp)=0
	Set the time stamp for this object in getTimeBase() units. More...

virtual IRational *	getTimeBase ()=0
	Get the time base that time stamps of this object are represented in. More...

virtual void	setTimeBase (IRational *aBase)=0
	Set the time base that time stamps of this object are represented in. More...

virtual com::avpkit::ferry::IBuffer *	getData ()=0
	Get any underlying raw data available for this object. More...

virtual bool	isKey ()=0
	Is this object a key object? i.e. More...

virtual void	setData (com::avpkit::ferry::IBuffer *buffer)=0
	Sets the underlying buffer used by this object. More...

Public Member Functions inherited from com::avpkit::ferry::RefCounted
virtual int32_t	acquire ()
	Internal Only. More...

virtual int32_t	release ()
	Internal Only. More...

virtual RefCounted *	copyReference ()
	Create a new Java object that refers to the same native object. More...

virtual int32_t	getCurrentRefCount ()
	Return the current reference count on this object. More...

void	setJavaAllocator (void *allocator)
	This method is public but not part of the standard API. More...

void *	getJavaAllocator ()
	This method is public but not part of the standard API. More...

Static Public Member Functions
static IVideoPicture *	make (IPixelFormat::Type format, int width, int height)
	Get a new picture object. More...

static IVideoPicture *	make (IVideoPicture *src)
	Get a new picture by copying the data in an existing frame. More...

static IVideoPicture *	make (com::avpkit::ferry::IBuffer *buffer, IPixelFormat::Type format, int width, int height)
	Get a new picture object, by wrapping an existing com.avpkit.ferry.IBuffer. More...

Additional Inherited Members
Protected Member Functions inherited from com::avpkit::ferry::RefCounted
virtual void	destroy ()
	This method is called by RefCounted objects when their Ref Count reaches zero and they are about to be destroyed.

Protected Attributes inherited from com::avpkit::ferry::RefCounted
AtomicInteger *	mRefCount
	This is the internal reference count, represented as an AtomicInteger to make sure it is thread safe.

void *	mAllocator
	Not part of public API.

Detailed Description

Represents one raw (undecoded) picture in a video stream, plus a timestamp for when to display that video picture relative to other items in a IContainer.

All timestamps for this object are always in Microseconds.

Definition at line 39 of file IVideoPicture.h.

Member Enumeration Documentation

◆ FrameDataType

enum com::avpkit::core::IVideoPicture::FrameDataType

Enumerator
AV_FRAME_DATA_PANSCAN	The data is the AVPanScan struct defined in libavcodec.
AV_FRAME_DATA_A53_CC	ATSC A53 Part 4 Closed Captions. A53 CC bitstream is stored as uint8_t in AVFrameSideData.data. The number of bytes of CC data is AVFrameSideData.size.
AV_FRAME_DATA_STEREO3D	Stereoscopic 3d metadata. The data is the AVStereo3D struct defined in libavutil/stereo3d.h.
AV_FRAME_DATA_MATRIXENCODING	The data is the AVMatrixEncoding enum defined in libavutil/channel_layout.h.
AV_FRAME_DATA_DOWNMIX_INFO	Metadata relevant to a downmix procedure. The data is the AVDownmixInfo struct defined in libavutil/downmix_info.h.
AV_FRAME_DATA_REPLAYGAIN	ReplayGain information in the form of the AVReplayGain struct.
AV_FRAME_DATA_DISPLAYMATRIX	This side data contains a 3x3 transformation matrix describing an affine transformation that needs to be applied to the frame for correct presentation. See libavutil/display.h for a detailed description of the data.
AV_FRAME_DATA_AFD	Active Format Description data consisting of a single byte as specified in ETSI TS 101 154 using AVActiveFormatDescription enum.
AV_FRAME_DATA_MOTION_VECTORS	Motion vectors exported by some codecs (on demand through the export_mvs flag set in the libavcodec AVCodecContext flags2 option). The data is the AVMotionVector struct defined in libavutil/motion_vector.h.
AV_FRAME_DATA_SKIP_SAMPLES	Recommmends skipping the specified number of samples. This is exported only if the "skip_manual" AVOption is set in libavcodec. This has the same format as AV_PKT_DATA_SKIP_SAMPLES. u32le number of samples to skip from start of this packet u32le number of samples to skip from end of this packet u8 reason for start skip u8 reason for end skip (0=padding silence, 1=convergence)
AV_FRAME_DATA_AUDIO_SERVICE_TYPE	This side data must be associated with an audio frame and corresponds to enum AVAudioServiceType defined in avcodec.h.
AV_FRAME_DATA_MASTERING_DISPLAY_METADATA	Mastering display metadata associated with a video frame. The payload is an AVMasteringDisplayMetadata type and contains information about the mastering display color volume.
AV_FRAME_DATA_GOP_TIMECODE	The GOP timecode in 25 bit timecode format. Data format is 64-bit integer. This is set on the first frame of a GOP that has a temporal reference of 0.
AV_FRAME_DATA_SPHERICAL	The data represents the AVSphericalMapping structure defined in libavutil/spherical.h.
AV_FRAME_DATA_CONTENT_LIGHT_LEVEL	Content light level (based on CTA-861.3). This payload contains data in the form of the AVContentLightMetadata struct.
AV_FRAME_DATA_ICC_PROFILE	The data contains an ICC profile as an opaque octet buffer following the format described by ISO 15076-1 with an optional name defined in the metadata key entry "name".

Definition at line 255 of file IVideoPicture.h.

                  {
         AV_FRAME_DATA_PANSCAN,
         AV_FRAME_DATA_A53_CC,
         AV_FRAME_DATA_STEREO3D,
         AV_FRAME_DATA_MATRIXENCODING,
         AV_FRAME_DATA_DOWNMIX_INFO,
         AV_FRAME_DATA_REPLAYGAIN,
         AV_FRAME_DATA_DISPLAYMATRIX,
         AV_FRAME_DATA_AFD,
         AV_FRAME_DATA_MOTION_VECTORS,
         AV_FRAME_DATA_SKIP_SAMPLES,
         AV_FRAME_DATA_AUDIO_SERVICE_TYPE,
         AV_FRAME_DATA_MASTERING_DISPLAY_METADATA,
         AV_FRAME_DATA_GOP_TIMECODE,
  
         AV_FRAME_DATA_SPHERICAL,
  
         AV_FRAME_DATA_CONTENT_LIGHT_LEVEL,
  
         AV_FRAME_DATA_ICC_PROFILE,
     } FrameDataType;

◆ PictType

enum com::avpkit::core::IVideoPicture::PictType

The different types of images that we can set.

See also: getPictureType()

Definition at line 221 of file IVideoPicture.h.

                  {
       DEFAULT_TYPE=0,
       I_TYPE = 1,
       P_TYPE = 2,
       B_TYPE = 3,
       S_TYPE = 4,
       SI_TYPE = 5,
       SP_TYPE = 6,
       BI_TYPE = 7,
     } PictType;

Member Function Documentation

◆ copy()

virtual bool com::avpkit::core::IVideoPicture::copy ( IVideoPicture * srcPicture )

pure virtual

Copy the contents of the given picture into this picture.

All buffers are copied by value, not be reference.

Parameters

srcPicture The picture you want to copy.

Returns: true if a successful copy; false if not.

Implemented in com::avpkit::core::VideoPicture.

Referenced by make().

◆ getDataLineSize()

virtual int com::avpkit::core::IVideoPicture::getDataLineSize ( int lineNo )

pure virtual

Return the size of each line in the VideoPicture data.

Usually there are no more than 4 lines, but the first line no that returns 0 is the end of the road.

Parameters

lineNo The line you want to know the (byte) size of.

Returns: The size (in bytes) of that line in data.

Implemented in com::avpkit::core::VideoPicture.

◆ getHeight()

virtual int com::avpkit::core::IVideoPicture::getHeight ( )

pure virtual

What is the height of the picture.

Returns: the height of the picture

Implemented in com::avpkit::core::VideoPicture.

Referenced by com::avpkit::core::VideoPicture::copy(), com::avpkit::core::IPixelFormat::getYUV420PPixelOffset(), and make().

◆ getPictureType()

virtual PictType com::avpkit::core::IVideoPicture::getPictureType ( )

pure virtual

Get the picture type.

This will be set on decoding to tell you what type of packet this was decoded from, and when encoding is a request to the encoder for how to encode the picture.

The request may be ignored by your codec.

Returns: the picture type.

Implemented in com::avpkit::core::VideoPicture.

◆ getPixelType()

virtual IPixelFormat::Type com::avpkit::core::IVideoPicture::getPixelType ( )

pure virtual

Returns the pixel format of the picture.

Returns: the pixel format of the picture.

Implemented in com::avpkit::core::VideoPicture.

Referenced by com::avpkit::core::VideoPicture::copy(), com::avpkit::core::IPixelFormat::getYUV420PPixelOffset(), and make().

◆ getPts()

virtual int64_t com::avpkit::core::IVideoPicture::getPts ( )

pure virtual

What is the Presentation Time Stamp (in Microseconds) of this picture.

The PTS is is scaled so that 1 PTS = 1/1,000,000 of a second.

Returns: the presentation time stamp (pts)

Implemented in com::avpkit::core::VideoPicture.

Referenced by com::avpkit::core::VideoPicture::copy().

◆ getQuality()

virtual int com::avpkit::core::IVideoPicture::getQuality ( )

pure virtual

This value is the quality setting this VideoPicture had when it was decoded, or is the value to use when this picture is next encoded (if reset with setQuality()

Returns: The quality.

Implemented in com::avpkit::core::VideoPicture.

◆ getSize()

virtual int32_t com::avpkit::core::IVideoPicture::getSize ( )

pure virtual

Total size in bytes of the decoded picture.

Returns: number of bytes of decoded picture

Implements com::avpkit::core::IMediaData.

Implemented in com::avpkit::core::VideoPicture.

◆ getWidth()

virtual int com::avpkit::core::IVideoPicture::getWidth ( )

pure virtual

What is the width of the picture.

Returns: the width of the picture

Implemented in com::avpkit::core::VideoPicture.

Referenced by com::avpkit::core::VideoPicture::copy(), com::avpkit::core::IPixelFormat::getYUV420PPixelOffset(), and make().

◆ isComplete()

virtual bool com::avpkit::core::IVideoPicture::isComplete ( )

pure virtual

Is this picture completely decoded?

Returns: is this picture completely decoded?

Implemented in com::avpkit::core::VideoPicture.

Referenced by com::avpkit::core::VideoPicture::copy().

◆ isKeyFrame()

virtual bool com::avpkit::core::IVideoPicture::isKeyFrame ( )

pure virtual

Is this a key frame?

Returns: is this a key frame

Implemented in com::avpkit::core::VideoPicture.

◆ make() [1/3]

IVideoPicture * com::avpkit::core::IVideoPicture::make	(	com::avpkit::ferry::IBuffer *	buffer,
		IPixelFormat::Type	format,
		int	width,
		int	height
	)

static

Get a new picture object, by wrapping an existing com.avpkit.ferry.IBuffer.

Use this method if you have existing video data that you want to have us wrap and pass to FFmpeg. Note that if decoding into this video picture and the decoded data actually takes more space than is in this buffer, this object will release the reference to the passed in buffer and allocate a new buffer instead so the decode can continue.

Due to some decoders assembly optimizations, you should ensure the IBuffer you pass in has at least 8 more bytes than would typically be required based on the format, width and height.

Parameters

buffer	The com.avpkit.ferry.IBuffer to wrap.
format	The pixel format (for example, YUV420P).
width	The width of the picture, in pixels.
height	The height of the picture, in pixels.

Returns: A new object, or null if we can't allocate one.

Definition at line 48 of file IVideoPicture.cpp.

   {
     Global::init();
     return VideoPicture::make(buffer, format, width, height);
   }

References com::avpkit::core::Global::init(), and com::avpkit::core::VideoPicture::make().

◆ make() [2/3]

IVideoPicture * com::avpkit::core::IVideoPicture::make	(	IPixelFormat::Type	format,
		int	width,
		int	height
	)

static

Get a new picture object.

You can specify -1 for width and height, in which case all getData() methods will return error until AVPKIT decodes something into this frame. In general you should always try to specify the width and height.

Note that any buffers this objects needs will be lazily allocated (i.e. we won't actually grab all the memory until we need it).

This is useful because it allows you to hold a IVideoPicture object that remembers things like format, width, and height, but know that it doesn't actually take up a lot of memory until the first time someone tries to access that memory.

Parameters

format	The pixel format (for example, YUV420P).
width	The width of the picture, in pixels, or -1 if you want AVPKIT to guess when decoding.
height	The height of the picture, in pixels, or -1 if you want AVPKIT to guess when decoding.

Returns: A new object, or null if we can't allocate one.

Definition at line 41 of file IVideoPicture.cpp.

   {
     Global::init();
     return VideoPicture::make(format, width, height);
   }

References com::avpkit::core::Global::init(), and com::avpkit::core::VideoPicture::make().

Referenced by make().

◆ make() [3/3]

IVideoPicture * com::avpkit::core::IVideoPicture::make ( IVideoPicture * src )

static

Get a new picture by copying the data in an existing frame.

Parameters

src	The picture to copy.

Returns: The new picture, or null on error.

Definition at line 58 of file IVideoPicture.cpp.

   {
     IVideoPicture* retval = 0;
     Global::init();
     try
     {
       if (!srcFrame)
         throw std::runtime_error("no source data to copy");
  
       retval = IVideoPicture::make(srcFrame->getPixelType(), srcFrame->getWidth(), srcFrame->getHeight());
       if (!retval)
         throw std::runtime_error("could not allocate new frame");
       
       if (!retval->copy(srcFrame))
         throw std::runtime_error("could not copy source frame");
     }
     catch (std::bad_alloc &e)
     {
       VS_REF_RELEASE(retval);
       throw e;
     }
     catch (std::exception & e)
     {
       VS_LOG_DEBUG("got error: %s", e.what());
       VS_REF_RELEASE(retval);
     }
     return retval;
   }

References copy(), getHeight(), getPixelType(), getWidth(), com::avpkit::core::Global::init(), and make().

◆ render()

virtual void com::avpkit::core::IVideoPicture::render	(	bool	drop = `false`,
		int64_t	timeStamp = `-1`
	)

pure virtual

Render this picture on configured surface.

Works only with HW accelerated com.avpkit.core.IPixelFormat. com.avpkit.core.IStreamCoder#setHardwareDecoding(IPixelFormat.Type, Object) must be called before opening decoder.

Implemented in com::avpkit::core::VideoPicture.

◆ setComplete()

virtual void com::avpkit::core::IVideoPicture::setComplete	(	bool	aIsComplete,
		IPixelFormat::Type	format,
		int	width,
		int	height,
		int64_t	pts
	)

pure virtual

After modifying the raw data in this buffer, call this function to let the object know it is now complete.

Parameters

aIsComplete	Is this VideoPicture complete
format	The pixel format of the data in this picture. Must match what the picture was originally constructed with.
width	The width of the data in this picture. Must match what the picture was originally constructed with.
height	The height of the data in this picture. Must match what the picture was originally constructed with.
pts	The presentation timestamp of the picture that is now complete. The caller must ensure this PTS is in units of 1/1,000,000 seconds.

Implemented in com::avpkit::core::VideoPicture.

◆ setKeyFrame()

virtual void com::avpkit::core::IVideoPicture::setKeyFrame ( bool aIsKey )

pure virtual

Reset if this is a key frame or not.

Note that regardless of how this flag is set, an IVideoPicture always contains raw video data (hence the key setting isn't really that important).

Parameters

aIsKey True if a key frame; false if not.

Implemented in com::avpkit::core::VideoPicture.

◆ setPictureType()

virtual void com::avpkit::core::IVideoPicture::setPictureType ( PictType type )

pure virtual

Set the picture type.

Parameters

type The type.

See also: getPictureType()

Implemented in com::avpkit::core::VideoPicture.

◆ setPts()

virtual void com::avpkit::core::IVideoPicture::setPts ( int64_t value )

pure virtual

Set the Presentation Time Stamp (in Microseconds) for this picture.

See also: getPts()

Parameters

value the new timestamp

Implemented in com::avpkit::core::VideoPicture.

◆ setQuality()

virtual void com::avpkit::core::IVideoPicture::setQuality ( int newQuality )

pure virtual

Set the Quality to a new value.

This will be used the next time this VideoPicture is encoded by a StreamCoder

Parameters

newQuality The new quality.

Implemented in com::avpkit::core::VideoPicture.

The documentation for this class was generated from the following files:

csrc/com/avpkit/core/IVideoPicture.h
csrc/com/avpkit/core/IVideoPicture.cpp

Public Types

Public Member Functions

Static Public Member Functions

Additional Inherited Members

Detailed Description

Member Enumeration Documentation

◆ FrameDataType

◆ PictType

Member Function Documentation

◆ copy()

◆ getDataLineSize()

◆ getHeight()

◆ getPictureType()

◆ getPixelType()

◆ getPts()

◆ getQuality()

◆ getSize()

◆ getWidth()

◆ isComplete()

◆ isKeyFrame()

◆ make() [1/3]

◆ make() [2/3]

◆ make() [3/3]

◆ render()

◆ setComplete()

◆ setKeyFrame()

◆ setPictureType()

◆ setPts()

◆ setQuality()