OpenCV  4.9.0
Open Source Computer Vision
Namespaces | Classes | Enumerations | Functions
cv::gapi::streaming Namespace Reference

This namespace contains G-API functions, structures, and symbols related to the Streaming execution mode. More...

Namespaces

 detail
 
 meta_tag
 

Classes

struct  queue_capacity
 Specify queue capacity for streaming execution. More...
 

Enumerations

enum  sync_policy {
  sync_policy::dont_sync,
  sync_policy::drop
}
 

Functions

cv::GMat BGR (const cv::GFrame &in)
 Gets bgr plane from input frame. More...
 
GMat desync (const GMat &g)
 Starts a desynchronized branch in the graph. More...
 
GFrame desync (const GFrame &f)
 
 G_API_OP (GY,< GMat(GFrame)>, "org.opencv.streaming.Y")
 
 G_API_OP (GUV,< GMat(GFrame)>, "org.opencv.streaming.UV")
 
 G_TYPED_KERNEL (GSize,< GOpaque< Size >(GMat)>, "org.opencv.streaming.size")
 
 G_TYPED_KERNEL (GSizeR,< GOpaque< Size >(GOpaque< Rect >)>, "org.opencv.streaming.sizeR")
 
 G_TYPED_KERNEL (GSizeMF,< GOpaque< Size >(GFrame)>, "org.opencv.streaming.sizeMF")
 
cv::GKernelPackage kernels ()
 Create a kernel package object containing kernels and transformations specified in variadic template argument. More...
 
template<typename T , typename G >
cv::GOpaque< T > meta (G g, const std::string &tag)
 
template<typename G >
cv::GOpaque< int64_tseq_id (G g)
 
template<typename G >
cv::GOpaque< int64_tseqNo (G g)
 
GOpaque< Sizesize (const GMat &src)
 Gets dimensions from Mat. More...
 
GOpaque< Sizesize (const GOpaque< Rect > &r)
 
GOpaque< Sizesize (const GFrame &src)
 Gets dimensions from MediaFrame. More...
 
template<typename G >
cv::GOpaque< int64_ttimestamp (G g)
 
GMat UV (const cv::GFrame &frame)
 Extracts UV plane from media frame. More...
 
GMat Y (const cv::GFrame &frame)
 Extracts Y plane from media frame. More...
 

Detailed Description

This namespace contains G-API functions, structures, and symbols related to the Streaming execution mode.

Some of the operations defined in this namespace (e.g. size(), BGR(), etc.) can be used in the traditional execution mode too.

Enumeration Type Documentation

◆ sync_policy

Enumerator
dont_sync 
drop 

Function Documentation

◆ BGR()

cv::GMat cv::gapi::streaming::BGR ( const cv::GFrame in)

Gets bgr plane from input frame.

Note
Function textual ID is "org.opencv.streaming.BGR"
Parameters
inInput frame
Returns
Image in BGR format

◆ desync() [1/2]

GMat cv::gapi::streaming::desync ( const GMat g)
Python:
cv.gapi.streaming.desync(g) -> retval

Starts a desynchronized branch in the graph.

This operation takes a single G-API data object and returns a graph-level "duplicate" of this object.

Operations which use this data object can be desynchronized from the rest of the graph.

This operation has no effect when a GComputation is compiled with regular cv::GComputation::compile(), since cv::GCompiled objects always produce their full output vectors.

This operation only makes sense when a GComputation is compiled in streaming mode with cv::GComputation::compileStreaming(). If this operation is used and there are desynchronized outputs, the user should use a special version of cv::GStreamingCompiled::pull() which produces an array of cv::util::optional<> objects.

Note
This feature is highly experimental now and is currently limited to a single GMat/GFrame argument only.

◆ desync() [2/2]

GFrame cv::gapi::streaming::desync ( const GFrame f)
Python:
cv.gapi.streaming.desync(g) -> retval

◆ G_API_OP() [1/2]

cv::gapi::streaming::G_API_OP ( GY  ,
< GMat(GFrame)>  ,
"org.opencv.streaming.Y"   
)

◆ G_API_OP() [2/2]

cv::gapi::streaming::G_API_OP ( GUV  ,
< GMat(GFrame)>  ,
"org.opencv.streaming.UV"   
)

◆ G_TYPED_KERNEL() [1/3]

cv::gapi::streaming::G_TYPED_KERNEL ( GSize  ,
< GOpaque< Size >(GMat)>  ,
"org.opencv.streaming.size"   
)

◆ G_TYPED_KERNEL() [2/3]

cv::gapi::streaming::G_TYPED_KERNEL ( GSizeR  ,
< GOpaque< Size >(GOpaque< Rect >)>  ,
"org.opencv.streaming.sizeR"   
)

◆ G_TYPED_KERNEL() [3/3]

cv::gapi::streaming::G_TYPED_KERNEL ( GSizeMF  ,
< GOpaque< Size >(GFrame)>  ,
"org.opencv.streaming.sizeMF"   
)

◆ meta()

template<typename T , typename G >
cv::GOpaque<T> cv::gapi::streaming::meta ( g,
const std::string &  tag 
)

◆ seq_id()

template<typename G >
cv::GOpaque<int64_t> cv::gapi::streaming::seq_id ( g)
Python:
cv.gapi.streaming.seq_id(arg1) -> retval

◆ seqNo()

template<typename G >
cv::GOpaque<int64_t> cv::gapi::streaming::seqNo ( g)
Python:
cv.gapi.streaming.seqNo(arg1) -> retval

◆ size() [1/3]

GOpaque<Size> cv::gapi::streaming::size ( const GMat src)
Python:
cv.gapi.streaming.size(src) -> retval
cv.gapi.streaming.size(r) -> retval

Gets dimensions from Mat.

Note
Function textual ID is "org.opencv.streaming.size"
Parameters
srcInput tensor
Returns
Size (tensor dimensions).
Examples:
fld_lines.cpp, samples/cpp/lkdemo.cpp, samples/cpp/stitching_detailed.cpp, samples/cpp/train_HOG.cpp, and samples/tapi/squares.cpp.

◆ size() [2/3]

GOpaque<Size> cv::gapi::streaming::size ( const GOpaque< Rect > &  r)
Python:
cv.gapi.streaming.size(src) -> retval
cv.gapi.streaming.size(r) -> retval

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts. Gets dimensions from rectangle.

Note
Function textual ID is "org.opencv.streaming.sizeR"
Parameters
rInput rectangle.
Returns
Size (rectangle dimensions).

◆ size() [3/3]

GOpaque<Size> cv::gapi::streaming::size ( const GFrame src)
Python:
cv.gapi.streaming.size(src) -> retval
cv.gapi.streaming.size(r) -> retval

Gets dimensions from MediaFrame.

Note
Function textual ID is "org.opencv.streaming.sizeMF"
Parameters
srcInput frame
Returns
Size (frame dimensions).

◆ timestamp()

template<typename G >
cv::GOpaque<int64_t> cv::gapi::streaming::timestamp ( g)
Python:
cv.gapi.streaming.timestamp(arg1) -> retval

◆ UV()

GMat cv::gapi::streaming::UV ( const cv::GFrame frame)

Extracts UV plane from media frame.

Output image is 8-bit 2-channel image of CV_8UC2.

Note
Function textual ID is "org.opencv.streaming.UV"
Parameters
frameinput media frame.

◆ Y()

GMat cv::gapi::streaming::Y ( const cv::GFrame frame)

Extracts Y plane from media frame.

Output image is 8-bit 1-channel image of CV_8UC1.

Note
Function textual ID is "org.opencv.streaming.Y"
Parameters
frameinput media frame.