Merge pull request #128 from AllenNeuralDynamics/fix-ffmpeg-gray-conv…

…ersion

Change gray online encoding string again

docs/articles/data_formats/video.rst

@@ -64,17 +64,23 @@ or
 - Acquire without any gamma correction
 - Use ``ffmpeg`` with the following encoding codec string for online encoding (optimized for compression quality and speed):
 
-  Note: this has been tested with monochrome videos with the raw pixel format
-  ``gray``. For color videos, the input arguments might need to be altered to
-  match the color space of the input.
-
-  - output arguments: ``-vf "scale=out_color_matrix=bt709:out_range=full" -c:v h264_nvenc -pix_fmt nv12 -color_range full -colorspace bt709 -color_trc linear -tune hq -preset p4 -rc vbr -cq 12 -b:v 0M -metadata author="Allen Institute for Neural Dynamics" -maxrate 700M -bufsize 350M``
+  .. note::
+     This pipeline has been designed and tested with monochrome videos with the raw pixel format ``gray``. For color videos, the arguments might need to be altered to match the color space of the input.
+
+  - output arguments: ``-vf "scale=out_color_matrix=bt709:out_range=full,format=bgr24,scale=out_range=full" -c:v h264_nvenc -pix_fmt yuv420p -color_range full -colorspace bt709 -color_trc linear -tune hq -preset p4 -rc vbr -cq 12 -b:v 0M -metadata author="Allen Institute for Neural Dynamics" -maxrate 700M -bufsize 350M``
   - input_arguments: ``-colorspace bt709 -color_primaries bt709 -color_range full -color_trc linear``
 
 and the following encoding codec string for offline re-encoding (optimized for quality and size):
 
 - output arguments: ``-vf "scale=out_color_matrix=bt709:out_range=full:sws_dither=none,format=yuv420p10le,colorspace=ispace=bt709:all=bt709:dither=none,scale=out_range=tv:sws_dither=none,format=yuv420p" -c:v libx264 -preset veryslow -crf 18 -pix_fmt yuv420p -metadata author="Allen Institute for Neural Dynamics" -movflags +faststart+write_colr``
 
+.. warning::
+
+  For higher bit depth (more than 8 bit) recordings, change the output arguments of the online, first stage, encoding to be as follows:
+    - output arguments: ``-vf "scale=out_color_matrix=bt709:out_range=full,format=rgb48le,scale=out_range=full" -c:v h264_nvenc -pix_fmt yuv420p -color_range full -colorspace bt709 -color_trc linear -tune hq -preset p4 -rc vbr -cq 12 -b:v 0M -metadata author="Allen Institute for Neural Dynamics" -maxrate 700M -bufsize 350M``
+
+This is almost the same, except the intermediate color representation is 48 bits per pixel instead of 24.
+
 Application notes
 #####################
 

src/aind_behavior_services/rig/__init__.py

@@ -21,11 +21,11 @@ class VideoWriterFfmpeg(BaseModel):
     frame_rate: int = Field(default=30, ge=0, description="Encoding frame rate")
     container_extension: str = Field(default="mp4", description="Container extension")
     output_arguments: str = Field(
-        default='-vf "scale=out_color_matrix=bt709:out_range=full" -c:v h264_nvenc -pix_fmt nv12 -color_range full -colorspace bt709 -color_trc linear -tune hq -preset p4 -rc vbr -cq 12 -b:v 0M -metadata author="Allen Institute for Neural Dynamics" -maxrate 700M -bufsize 350M',  # E501
+        default='-vf "scale=out_color_matrix=bt709:out_range=full,format=bgr24,scale=out_range=full" -c:v h264_nvenc -pix_fmt yuv420p -color_range full -colorspace bt709 -color_trc linear -tune hq -preset p4 -rc vbr -cq 12 -b:v 0M -metadata author="Allen Institute for Neural Dynamics" -maxrate 700M -bufsize 350M',  # E501
         description="Output arguments",
     )
     input_arguments: str = Field(
-        default="-v verbose -colorspace bt709 -color_primaries bt709 -color_range full -color_trc linear",
+        default="-colorspace bt709 -color_primaries bt709 -color_range full -color_trc linear",
         description="Input arguments",
     )