A recent obs-deps change removed all non-essential x86 deps. This caused
the x86 subproject(s) on Windows to fail to configure due to being
unable to find x86 dependencies that do not exist.
Co-authored-by: PatTheMav <PatTheMav@users.noreply.github.com>
The code prior to this change would never add virtualcam.c to the
win-dshow target, which resulted in the virtualcam_output not being
registered and thus Virtual Camera not working.
Co-authored-by: PatTheMav <PatTheMav@users.noreply.github.com>
This event is only used within destroy_[audio]_screen_stream, and does
not appear to be necessary there. stopCaptureWithCompletionHandler holds
a reference to the SCStream object by itself (and the other objects
being held aren't used afterwards anyways), so there should be no harm
in releasing everything immediately without blocking.
This adds a Background Blur filter which blurs the background for a
foreground speaker identified by NVIDIA AI Greenscreen FX.
Secondly, this adds a Blur filter based on NVIDIA Video Effects,
leveraging the Background Blur Filter where the mask used just sets the
whole frame as background.
Signed-off-by: pkv <pkv@obsproject.com>
Adds a new encoder cap which tells libobs that rather than scaling
video frames in software, the encoder is capable of scaling them via
its own (presumably more efficient) means.
An encoder may implement this cap by comparing the VOI of its assigned
`video_t` and the results of `obs_encoder_get_width/height()`. If the
width/height values differ, then the encoder is being asked by libobs
to self-scale, and the resolution in VOI will be the raw frame size,
with the `...get_width/height()` being the intended output resolution
of the encoder.
It is important to note that GPU rescaling mode will take priority
over self-scaling. If GPU rescaling is enabled, the encoder will never
be asked to self-scale.
This is useful for discrete hardware encoders, where they might have
fixed-function video scaling logic that is highly efficient and fast.
Additionally, this feature allows a hardware device which is encoding
a full ABR ladder of tracks to be smart and only copy a video frame
from GPU -> Host -> Device once for the entire ladder, rather than
once for every track.
This logic would previously have written any changed scale resolution
set by the encoder in the `.get_video_info` callback back to the
encoder, however this functionality was "broken" by
20d8779d30. In reality, this would have
never worked with texture encoders or with GPU rescaling enabled, and
probably would have had odd side effects for CPU rescaling, too. It's
best just to remove this functionality.
When a source file contains an explicit include with a filename
following the "moc_<actual-filename>.cpp" pattern, then CMake's
AUTOMOC generation tool will recognize the matching pair and generate
the replacement header file and add the required include directory
entries.
For all files which do contain Q_OBJECT or similar declarations but do
not have an explicit include directive, the global mocs_compilation.cpp
file will still be generated (which groups all "missing" generated
headers).
The larger this global file is, the more expensive incremental
compilation will be as this file (and all its contained generated
headers) will be re-generated regardless of whether actual changes
occurred.
When a source file contains an explicit include with a filename
following the "moc_<actual-filename>.cpp" pattern, then CMake's
AUTOMOC generation tool will recognize the matching pair and generate
the replacement header file and add the required include directory
entries.
For all files which do contain Q_OBJECT or similar declarations but do
not have an explicit include directive, the global mocs_compilation.cpp
file will still be generated (which groups all "missing" generated
headers).
The larger this global file is, the more expensive incremental
compilation will be as this file (and all its contained generated
headers) will be re-generated regardless of whether actual changes
occurred.
* Update libvpl to version 2.12.0
* Update vpl-gpu-rt to version 24.2.5
* 24.3 requires VVC from libva which is not the case with the actual
runtime in use
* Update nv-codec to version 12.2.72.0
The ComPtr, WinHandle, and obfuscate source files are shared with
multiple sub-projects and thus need to be moved into the shared
directory to uncouple their availability from libobs itself.
The same applies to d3d8-api, inject-library, and hook-config
(from win-capture), as well as comutils (from the virtualcam-module).
CMake doesn't expect CMake package files to be separated in
subdirectories on Windows and instead expects all files to be put in
a single directory `cmake` found within one of the PREFIX paths.
Also fixes circular dependency in w32-pthreads CMake package config
file.
