gstreamer-rs/docs/gstreamer-video/docs.md
2017-08-11 15:37:32 +03:00

32 KiB

The color matrix is used to convert between Y'PbPr and non-linear RGB (R'G'B')

unknown matrix

identity matrix

FCC color matrix

ITU-R BT.709 color matrix

ITU-R BT.601 color matrix

SMPTE 240M color matrix

ITU-R BT.2020 color matrix. Since: 1.6.

The color primaries define the how to transform linear RGB values to and from the CIE XYZ colorspace.

unknown color primaries

BT709 primaries

BT470M primaries

BT470BG primaries

SMPTE170M primaries

SMPTE240M primaries

Generic film

BT2020 primaries. Since: 1.6.

Adobe RGB primaries. Since: 1.8

Possible color range values. These constants are defined for 8 bit color values and can be scaled for other bit depths.

unknown range

[0..255] for 8 bit components

[16..235] for 8 bit components. Chroma has [16..240] range.

Structure describing the color info.

Parse the colorimetry string and update self with the parsed values.

color

a colorimetry string

Returns

true if color points to valid colorimetry info.

Compare the 2 colorimetry sets for equality

other

another VideoColorimetry

Returns

true if self and other are equal.

Check if the colorimetry information in info matches that of the string color.

color

a colorimetry string

Returns

true if color conveys the same colorimetry info as the color information in info.

Make a string representation of self.

Returns

a string representation of self.

Field order of interlaced content. This is only valid for interlace-mode=interleaved and not interlace-mode=mixed. In the case of mixed or GST_VIDEO_FIELD_ORDER_UNKOWN, the field order is signalled via buffer flags.

unknown field order for interlaced content. The actual field order is signalled via buffer flags.

top field is first

bottom field is first

Feature: v1_12

Enum value describing the most common video formats.

Unknown or unset video format id

Encoded video format. Only ever use that in caps for special video formats in combination with non-system memory GstCapsFeatures where it does not make sense to specify a real video format.

planar 4:2:0 YUV

planar 4:2:0 YVU (like I420 but UV planes swapped)

packed 4:2:2 YUV (Y0-U0-Y1-V0 Y2-U2-Y3-V2 Y4 ...)

packed 4:2:2 YUV (U0-Y0-V0-Y1 U2-Y2-V2-Y3 U4 ...)

packed 4:4:4 YUV with alpha channel (A0-Y0-U0-V0 ...)

sparse rgb packed into 32 bit, space last

sparse reverse rgb packed into 32 bit, space last

sparse rgb packed into 32 bit, space first

sparse reverse rgb packed into 32 bit, space first

rgb with alpha channel last

reverse rgb with alpha channel last

rgb with alpha channel first

reverse rgb with alpha channel first

rgb

reverse rgb

planar 4:1:1 YUV

planar 4:2:2 YUV

packed 4:2:2 YUV (Y0-V0-Y1-U0 Y2-V2-Y3-U2 Y4 ...)

planar 4:4:4 YUV

packed 4:2:2 10-bit YUV, complex format

packed 4:2:2 16-bit YUV, Y0-U0-Y1-V1 order

planar 4:2:0 YUV with interleaved UV plane

planar 4:2:0 YUV with interleaved VU plane

8-bit grayscale

16-bit grayscale, most significant byte first

16-bit grayscale, least significant byte first

packed 4:4:4 YUV (Y-U-V ...)

rgb 5-6-5 bits per component

reverse rgb 5-6-5 bits per component

rgb 5-5-5 bits per component

reverse rgb 5-5-5 bits per component

packed 10-bit 4:2:2 YUV (U0-Y0-V0-Y1 U2-Y2-V2-Y3 U4 ...)

planar 4:4:2:0 AYUV

8-bit paletted RGB

planar 4:1:0 YUV

planar 4:1:0 YUV (like YUV9 but UV planes swapped)

packed 4:1:1 YUV (Cb-Y0-Y1-Cr-Y2-Y3 ...)

rgb with alpha channel first, 16 bits per channel

packed 4:4:4 YUV with alpha channel, 16 bits per channel (A0-Y0-U0-V0 ...)

packed 4:4:4 RGB, 10 bits per channel

planar 4:2:0 YUV, 10 bits per channel

planar 4:2:0 YUV, 10 bits per channel

planar 4:2:2 YUV, 10 bits per channel

planar 4:2:2 YUV, 10 bits per channel

planar 4:4:4 YUV, 10 bits per channel (Since: 1.2)

planar 4:4:4 YUV, 10 bits per channel (Since: 1.2)

planar 4:4:4 RGB, 8 bits per channel (Since: 1.2)

planar 4:4:4 RGB, 10 bits per channel (Since: 1.2)

planar 4:4:4 RGB, 10 bits per channel (Since: 1.2)

planar 4:2:2 YUV with interleaved UV plane (Since: 1.2)

planar 4:4:4 YUV with interleaved UV plane (Since: 1.2)

NV12 with 64x32 tiling in zigzag pattern (Since: 1.4)

planar 4:4:2:0 YUV, 10 bits per channel (Since: 1.6)

planar 4:4:2:0 YUV, 10 bits per channel (Since: 1.6)

planar 4:4:2:2 YUV, 10 bits per channel (Since: 1.6)

planar 4:4:2:2 YUV, 10 bits per channel (Since: 1.6)

planar 4:4:4:4 YUV, 10 bits per channel (Since: 1.6)

planar 4:4:4:4 YUV, 10 bits per channel (Since: 1.6)

planar 4:2:2 YUV with interleaved VU plane (Since: 1.6)

planar 4:2:0 YUV with interleaved UV plane, 10 bits per channel (Since: 1.10)

planar 4:2:0 YUV with interleaved UV plane, 10 bits per channel (Since: 1.10)

packed 4:4:4 YUV (U-Y-V ...) (Since 1.10)

packed 4:2:2 YUV (V0-Y0-U0-Y1 V2-Y2-U2-Y3 V4 ...)

planar 4:4:4:4 ARGB, 8 bits per channel (Since: 1.12)

planar 4:4:4:4 ARGB, 10 bits per channel (Since: 1.12)

planar 4:4:4:4 ARGB, 10 bits per channel (Since: 1.12)

planar 4:4:4 RGB, 12 bits per channel (Since: 1.12)

planar 4:4:4 RGB, 12 bits per channel (Since: 1.12)

planar 4:4:4:4 ARGB, 12 bits per channel (Since: 1.12)

planar 4:4:4:4 ARGB, 12 bits per channel (Since: 1.12)

planar 4:2:0 YUV, 12 bits per channel (Since: 1.12)

planar 4:2:0 YUV, 12 bits per channel (Since: 1.12)

planar 4:2:2 YUV, 12 bits per channel (Since: 1.12)

planar 4:2:2 YUV, 12 bits per channel (Since: 1.12)

planar 4:4:4 YUV, 12 bits per channel (Since: 1.12)

planar 4:4:4 YUV, 12 bits per channel (Since: 1.12)

Information for a video format.

A video frame obtained from VideoFrame::map

Copy the contents from src to self.

src

a VideoFrame

Returns

TRUE if the contents could be copied.

Copy the plane with index plane from src to self.

src

a VideoFrame

plane

a plane

Returns

TRUE if the contents could be copied.

Use info and buffer to fill in the values of self. self is usually allocated on the stack, and you will pass the address to the VideoFrame structure allocated on the stack; VideoFrame::map will then fill in the structures with the various video-specific information you need to access the pixels of the video buffer. You can then use accessor macros such as GST_VIDEO_FRAME_COMP_DATA(), GST_VIDEO_FRAME_PLANE_DATA(), GST_VIDEO_FRAME_COMP_STRIDE(), GST_VIDEO_FRAME_PLANE_STRIDE() etc. to get to the pixels.

  GstVideoFrame vframe;
  ...
  // set RGB pixels to black one at a time
  if (gst_video_frame_map (&vframe, video_info, video_buffer, GST_MAP_WRITE)) {
    guint8 *pixels = GST_VIDEO_FRAME_PLANE_DATA (vframe, 0);
    guint stride = GST_VIDEO_FRAME_PLANE_STRIDE (vframe, 0);
    guint pixel_stride = GST_VIDEO_FRAME_COMP_PSTRIDE (vframe, 0);

    for (h = 0; h < height; ++h) {
      for (w = 0; w < width; ++w) {
        guint8 *pixel = pixels + h * stride + w * pixel_stride;

        memset (pixel, 0, pixel_stride);
      }
    }

    gst_video_frame_unmap (&amp;vframe);
  }
  ...

All video planes of buffer will be mapped and the pointers will be set in self->data.

The purpose of this function is to make it easy for you to get to the video pixels in a generic way, without you having to worry too much about details such as whether the video data is allocated in one contiguous memory chunk or multiple memory chunks (e.g. one for each plane); or if custom strides and custom plane offsets are used or not (as signalled by GstVideoMeta on each buffer). This function will just fill the VideoFrame structure with the right values and if you use the accessor macros everything will just work and you can access the data easily. It also maps the underlying memory chunks for you.

info

a VideoInfo

buffer

the buffer to map

flags

gst::MapFlags

Returns

true on success.

Use info and buffer to fill in the values of self with the video frame information of frame id.

When id is -1, the default frame is mapped. When id != -1, this function will return false when there is no GstVideoMeta with that id.

All video planes of buffer will be mapped and the pointers will be set in self->data.

info

a VideoInfo

buffer

the buffer to map

id

the frame id to map

flags

gst::MapFlags

Returns

true on success.

Unmap the memory previously mapped with gst_video_frame_map.

Information describing image properties. This information can be filled in from GstCaps with VideoInfo::from_caps. The information is also used to store the specific video info when mapping a video frame with VideoFrame::map.

Use the provided macros to access the info in this structure.

Allocate a new VideoInfo that is also initialized with VideoInfo::init.

Returns

a new VideoInfo. free with VideoInfo::free.

Adjust the offset and stride fields in self so that the padding and stride alignment in align is respected.

Extra padding will be added to the right side when stride alignment padding is required and align will be updated with the new padding values.

align

alignment parameters

Returns

false if alignment could not be applied, e.g. because the size of a frame can't be represented as a 32 bit integer (Since: 1.12)

Converts among various gst::Format types. This function handles GST_FORMAT_BYTES, GST_FORMAT_TIME, and GST_FORMAT_DEFAULT. For raw video, GST_FORMAT_DEFAULT corresponds to video frames. This function can be used to handle pad queries of the type GST_QUERY_CONVERT.

src_format

gst::Format of the src_value

src_value

value to convert

dest_format

gst::Format of the dest_value

dest_value

pointer to destination value

Returns

TRUE if the conversion was successful.

Copy a GstVideoInfo structure.

Returns

a new VideoInfo. free with gst_video_info_free.

Free a GstVideoInfo structure previously allocated with VideoInfo::new or VideoInfo::copy.

Parse caps and update self.

caps

a gst::Caps

Returns

TRUE if caps could be parsed

Initialize self with default values.

Compares two VideoInfo and returns whether they are equal or not

other

a VideoInfo

Returns

true if self and other are equal, else false.

Set the default info for a video frame of format and width and height.

Note: This initializes self first, no values are preserved. This function does not set the offsets correctly for interlaced vertically subsampled formats.

format

the format

width

a width

height

a height

Returns

false if the returned video info is invalid, e.g. because the size of a frame can't be represented as a 32 bit integer (Since: 1.12)

Convert the values of self into a gst::Caps.

Returns

a new gst::Caps containing the info of self.

The possible values of the VideoInterlaceMode describing the interlace mode of the stream.

all frames are progressive

2 fields are interleaved in one video frame. Extra buffer flags describe the field order.

frames contains both interlaced and progressive video, the buffer flags describe the frame and fields.

2 fields are stored in one buffer, use the frame ID to get access to the required field. For multiview (the 'views' property > 1) the fields of view N can be found at frame ID (N * 2) and (N * 2) + 1. Each field has only half the amount of lines as noted in the height property. This mode requires multiple GstVideoMeta metadata to describe the fields.

All possible stereoscopic 3D and multiview representations. In conjunction with VideoMultiviewFlags, describes how multiview content is being transported in the stream.

A special value indicating no multiview information. Used in GstVideoInfo and other places to indicate that no specific multiview handling has been requested or provided. This value is never carried on caps.

All frames are monoscopic.

All frames represent a left-eye view.

All frames represent a right-eye view.

Left and right eye views are provided in the left and right half of the frame respectively.

Left and right eye views are provided in the left and right half of the frame, but have been sampled using quincunx method, with half-pixel offset between the 2 views.

Alternating vertical columns of pixels represent the left and right eye view respectively.

Alternating horizontal rows of pixels represent the left and right eye view respectively.

The top half of the frame contains the left eye, and the bottom half the right eye.

Pixels are arranged with alternating pixels representing left and right eye views in a checkerboard fashion.

Left and right eye views are provided in separate frames alternately.

Multiple independent views are provided in separate frames in sequence. This method only applies to raw video buffers at the moment. Specific view identification is via the GstVideoMultiviewMeta and VideoMeta(s) on raw video buffers.

Multiple views are provided as separate gst::Memory framebuffers attached to each gst::Buffer, described by the GstVideoMultiviewMeta and VideoMeta(s)

The VideoOverlay interface is used for 2 main purposes :

  • To get a grab on the Window where the video sink element is going to render. This is achieved by either being informed about the Window identifier that the video sink element generated, or by forcing the video sink element to use a specific Window identifier for rendering.
  • To force a redrawing of the latest video frame the video sink element displayed on the Window. Indeed if the gst::Pipeline is in gst::State::Paused state, moving the Window around will damage its content. Application developers will want to handle the Expose events themselves and force the video sink element to refresh the Window's content.

Using the Window created by the video sink is probably the simplest scenario, in some cases, though, it might not be flexible enough for application developers if they need to catch events such as mouse moves and button clicks.

Setting a specific Window identifier on the video sink element is the most flexible solution but it has some issues. Indeed the application needs to set its Window identifier at the right time to avoid internal Window creation from the video sink element. To solve this issue a gst::Message is posted on the bus to inform the application that it should set the Window identifier immediately. Here is an example on how to do that correctly:

static GstBusSyncReply
create_window (GstBus * bus, GstMessage * message, GstPipeline * pipeline)
{
 // ignore anything but 'prepare-window-handle' element messages
 if (!gst_is_video_overlay_prepare_window_handle_message (message))
   return GST_BUS_PASS;

 win = XCreateSimpleWindow (disp, root, 0, 0, 320, 240, 0, 0, 0);

 XSetWindowBackgroundPixmap (disp, win, None);

 XMapRaised (disp, win);

 XSync (disp, FALSE);

 gst_video_overlay_set_window_handle (GST_VIDEO_OVERLAY (GST_MESSAGE_SRC (message)),
     win);

 gst_message_unref (message);

 return GST_BUS_DROP;
}
...
int
main (int argc, char **argv)
{
...
 bus = gst_pipeline_get_bus (GST_PIPELINE (pipeline));
 gst_bus_set_sync_handler (bus, (GstBusSyncHandler) create_window, pipeline,
        NULL);
...
}

Two basic usage scenarios

There are two basic usage scenarios: in the simplest case, the application uses playbin or plasink or knows exactly what particular element is used for video output, which is usually the case when the application creates the videosink to use (e.g. xvimagesink, ximagesink, etc.) itself; in this case, the application can just create the videosink element, create and realize the window to render the video on and then call VideoOverlay::set_window_handle directly with the XID or native window handle, before starting up the pipeline. As playbin and playsink implement the video overlay interface and proxy it transparently to the actual video sink even if it is created later, this case also applies when using these elements.

In the other and more common case, the application does not know in advance what GStreamer video sink element will be used for video output. This is usually the case when an element such as autovideosink is used. In this case, the video sink element itself is created asynchronously from a GStreamer streaming thread some time after the pipeline has been started up. When that happens, however, the video sink will need to know right then whether to render onto an already existing application window or whether to create its own window. This is when it posts a prepare-window-handle message, and that is also why this message needs to be handled in a sync bus handler which will be called from the streaming thread directly (because the video sink will need an answer right then).

As response to the prepare-window-handle element message in the bus sync handler, the application may use VideoOverlay::set_window_handle to tell the video sink to render onto an existing window surface. At this point the application should already have obtained the window handle / XID, so it just needs to set it. It is generally not advisable to call any GUI toolkit functions or window system functions from the streaming thread in which the prepare-window-handle message is handled, because most GUI toolkits and windowing systems are not thread-safe at all and a lot of care would be required to co-ordinate the toolkit and window system calls of the different threads (Gtk+ users please note: prior to Gtk+ 2.18 GDK_WINDOW_XID() was just a simple structure access, so generally fine to do within the bus sync handler; this macro was changed to a function call in Gtk+ 2.18 and later, which is likely to cause problems when called from a sync handler; see below for a better approach without GDK_WINDOW_XID() used in the callback).

GstVideoOverlay and Gtk+

#include &lt;gst/video/videooverlay.h&gt;
#include &lt;gtk/gtk.h&gt;
#ifdef GDK_WINDOWING_X11
#include &lt;gdk/gdkx.h&gt;  // for GDK_WINDOW_XID
#endif
#ifdef GDK_WINDOWING_WIN32
#include &lt;gdk/gdkwin32.h&gt;  // for GDK_WINDOW_HWND
#endif
...
static guintptr video_window_handle = 0;
...
static GstBusSyncReply
bus_sync_handler (GstBus * bus, GstMessage * message, gpointer user_data)
{
 // ignore anything but 'prepare-window-handle' element messages
 if (!gst_is_video_overlay_prepare_window_handle_message (message))
   return GST_BUS_PASS;

 if (video_window_handle != 0) {
   GstVideoOverlay *overlay;

   // GST_MESSAGE_SRC (message) will be the video sink element
   overlay = GST_VIDEO_OVERLAY (GST_MESSAGE_SRC (message));
   gst_video_overlay_set_window_handle (overlay, video_window_handle);
 } else {
   g_warning ("Should have obtained video_window_handle by now!");
 }

 gst_message_unref (message);
 return GST_BUS_DROP;
}
...
static void
video_widget_realize_cb (GtkWidget * widget, gpointer data)
{
#if GTK_CHECK_VERSION(2,18,0)
  // Tell Gtk+/Gdk to create a native window for this widget instead of
  // drawing onto the parent widget.
  // This is here just for pedagogical purposes, GDK_WINDOW_XID will call
  // it as well in newer Gtk versions
  if (!gdk_window_ensure_native (widget->window))
    g_error ("Couldn't create native window needed for GstVideoOverlay!");
#endif

#ifdef GDK_WINDOWING_X11
  {
    gulong xid = GDK_WINDOW_XID (gtk_widget_get_window (video_window));
    video_window_handle = xid;
  }
#endif
#ifdef GDK_WINDOWING_WIN32
  {
    HWND wnd = GDK_WINDOW_HWND (gtk_widget_get_window (video_window));
    video_window_handle = (guintptr) wnd;
  }
#endif
}
...
int
main (int argc, char **argv)
{
  GtkWidget *video_window;
  GtkWidget *app_window;
  ...
  app_window = gtk_window_new (GTK_WINDOW_TOPLEVEL);
  ...
  video_window = gtk_drawing_area_new ();
  g_signal_connect (video_window, "realize",
      G_CALLBACK (video_widget_realize_cb), NULL);
  gtk_widget_set_double_buffered (video_window, FALSE);
  ...
  // usually the video_window will not be directly embedded into the
  // application window like this, but there will be many other widgets
  // and the video window will be embedded in one of them instead
  gtk_container_add (GTK_CONTAINER (ap_window), video_window);
  ...
  // show the GUI
  gtk_widget_show_all (app_window);

  // realize window now so that the video window gets created and we can
  // obtain its XID/HWND before the pipeline is started up and the videosink
  // asks for the XID/HWND of the window to render onto
  gtk_widget_realize (video_window);

  // we should have the XID/HWND now
  g_assert (video_window_handle != 0);
  ...
  // set up sync handler for setting the xid once the pipeline is started
  bus = gst_pipeline_get_bus (GST_PIPELINE (pipeline));
  gst_bus_set_sync_handler (bus, (GstBusSyncHandler) bus_sync_handler, NULL,
      NULL);
  gst_object_unref (bus);
  ...
  gst_element_set_state (pipeline, GST_STATE_PLAYING);
  ...
}

GstVideoOverlay and Qt

#include &lt;glib.h&gt;
#include &lt;gst/gst.h&gt;
#include &lt;gst/video/videooverlay.h&gt;

#include &lt;QApplication&gt;
#include &lt;QTimer&gt;
#include &lt;QWidget&gt;

int main(int argc, char *argv[])
{
  if (!g_thread_supported ())
    g_thread_init (NULL);

  gst_init (&argc, &argv);
  QApplication app(argc, argv);
  app.connect(&app, SIGNAL(lastWindowClosed()), &app, SLOT(quit ()));

  // prepare the pipeline

  GstElement *pipeline = gst_pipeline_new ("xvoverlay");
  GstElement *src = gst_element_factory_make ("videotestsrc", NULL);
  GstElement *sink = gst_element_factory_make ("xvimagesink", NULL);
  gst_bin_add_many (GST_BIN (pipeline), src, sink, NULL);
  gst_element_link (src, sink);

  // prepare the ui

  QWidget window;
  window.resize(320, 240);
  window.show();

  WId xwinid = window.winId();
  gst_video_overlay_set_window_handle (GST_VIDEO_OVERLAY (sink), xwinid);

  // run the pipeline

  GstStateChangeReturn sret = gst_element_set_state (pipeline,
      GST_STATE_PLAYING);
  if (sret == GST_STATE_CHANGE_FAILURE) {
    gst_element_set_state (pipeline, GST_STATE_NULL);
    gst_object_unref (pipeline);
    // Exit application
    QTimer::singleShot(0, QApplication::activeWindow(), SLOT(quit()));
  }

  int ret = app.exec();

  window.hide();
  gst_element_set_state (pipeline, GST_STATE_NULL);
  gst_object_unref (pipeline);

  return ret;
}

Implements

VideoOverlayExt

Trait containing all VideoOverlay methods.

Implementors

VideoOverlay

Tell an overlay that it has been exposed. This will redraw the current frame in the drawable even if the pipeline is PAUSED.

This will post a "have-window-handle" element message on the bus.

This function should only be used by video overlay plugin developers.

handle

a platform-specific handle referencing the window

Tell an overlay that it should handle events from the window system. These events are forwarded upstream as navigation events. In some window system, events are not propagated in the window hierarchy if a client is listening for them. This method allows you to disable events handling completely from the VideoOverlay.

handle_events

a gboolean indicating if events should be handled or not.

This will post a "prepare-window-handle" element message on the bus to give applications an opportunity to call VideoOverlay::set_window_handle before a plugin creates its own window.

This function should only be used by video overlay plugin developers.

Configure a subregion as a video target within the window set by VideoOverlay::set_window_handle. If this is not used or not supported the video will fill the area of the window set as the overlay to 100%. By specifying the rectangle, the video can be overlayed to a specific region of that window only. After setting the new rectangle one should call VideoOverlay::expose to force a redraw. To unset the region pass -1 for the width and height parameters.

This method is needed for non fullscreen video overlay in UI toolkits that do not support subwindows.

x

the horizontal offset of the render area inside the window

y

the vertical offset of the render area inside the window

width

the width of the render area inside the window

height

the height of the render area inside the window

Returns

false if not supported by the sink.

This will call the video overlay's set_window_handle method. You should use this method to tell to an overlay to display video output to a specific window (e.g. an XWindow on X11). Passing 0 as the handle will tell the overlay to stop using that window and create an internal one.

handle

a handle referencing the window.

Enum value describing the available tiling modes.

Unknown or unset tile mode

Every four adjacent blocks - two horizontally and two vertically are grouped together and are located in memory in Z or flipped Z order. In case of odd rows, the last row of blocks is arranged in linear order.

The video transfer function defines the formula for converting between non-linear RGB (R'G'B') and linear RGB

unknown transfer function

linear RGB, gamma 1.0 curve

Gamma 1.8 curve

Gamma 2.0 curve

Gamma 2.2 curve

Gamma 2.2 curve with a linear segment in the lower range

Gamma 2.2 curve with a linear segment in the lower range

Gamma 2.4 curve with a linear segment in the lower range

Gamma 2.8 curve

Logarithmic transfer characteristic 100:1 range

Logarithmic transfer characteristic 316.22777:1 range

Gamma 2.2 curve with a linear segment in the lower range. Used for BT.2020 with 12 bits per component. Since: 1.6.

Gamma 2.19921875. Since: 1.8