vcdp / ffmpeg-patch Goto Github PK

Hello again,

I switched to Ubuntu 16.04 and tried to apply the patches to ffmpeg and the patches don't seem to be applied.

The system is clean as I just installed it over the weekend.

Following Guidance 3.2 Build on a host machine :
ALL the required dependencies are successfully installed through apt-get install.
I used OpenVino_p20.2.120 version. Excluded Intel-media-sdk. Did not see OpenVX in the optionally-excluded list.

libdrm, libva, & media-sdk, etc are properly installed
Clinfo & Vainfo run just fine.
Patching process runs without a hitch.

In the final building stage, I used build-ffmpeg-dynamic and disabled librkafka as there is always an error when building its file.

At last, when I checked the installation with ffmepg -filters | grep Inference command, the output only showed ffmpeg default filters without "detect" & "classify".

Any suggestions ?

Hi,

Can you share with us the application where this dldt-c-api is used for easy understanding of the benefit of it.

Hi, there

I am stuck at installation process 3.2 Installation of Intel Media SDK
--> vainfo can't find iHD driver.
My cpu is E5 2620 v4 @ 2.10Ghz

The problem might be that it was a server cpu and there is no iGPU (so no available driver ?).
Is there a way to bypass that and use Cuda to for video acceleration ?

I would suggest to add SRT protocol to the docker image:

## libsrt https://github.com/Haivision/srt ARG LIBSRT_VERSION=1.4.1 ENV DEBIAN_FRONTEND noninteractive RUN apt-get install libssl-dev tclsh -y RUN DIR=/tmp/srt && \ mkdir -p ${DIR} && \ cd ${DIR} && \ wget https://github.com/Haivision/srt/archive/v${LIBSRT_VERSION}.tar.gz && \ tar -xz --strip-components=1 -f v${LIBSRT_VERSION}.tar.gz && \ cmake -DCMAKE_INSTALL_PREFIX=/usr/local && \ make && \ make install DESTDIR="/home/build" && \ make install && \ rm -rf ${DIR}

Then compile ffmpeg with this ability:

RUN cd /home/FFmpeg && \ find ${FFMPEG_MA_PATH}/patches -type f -name '*.patch' -print0 | sort -z | xargs -t -0 -n 1 patch -p1 -i && \ cp /root/patch/opencv.pc /usr/lib/pkgconfig/ && \ cp /root/patch/cvdef.h /usr/local/include/opencv4/opencv2/core/cvdef.h && \ ./configure --prefix="/usr" --extra-cflags="-I$IE_PATH/include" --extra-ldflags="-L$IE_PATH/lib/intel64" --libdir=/usr/lib/x86_64-linux-gnu --extra-libs="-lpthread -lm" --enable-shared --enable-gpl --enable-libass --enable-libfreetype --enable-openssl --enable-nonfree --enable-libdrm --enable-libmfx --disable-doc --disable-htmlpages --disable-manpages --disable-podpages --disable-txtpages --enable-libfdk-aac --enable-libmp3lame --enable-libopus --enable-libvorbis --enable-libx264 --enable-libx265 --enable-libjson_c --enable-libsrt --enable-libinference_engine_c_api --enable-libopencv --enable-python3 && \ make -j8 && \ make install DESTDIR="/home/build" && \ cp -rf python /home/build/python

Thanks! :)

Hi

When I run this command.
ffmpeg -i test.mp4 -vf detect=model=face-detection-adas-0001.xml:model_proc=face-detection-adas-0001.json,classify=model=emotions-recognition-retail-0003.xml:model_proc=emotions-recognition-retail-0003.json -an -f null -

It shows like this.
[AVFilterGraph @ 0x55de3f70a200] No such filter: 'detect'
Error reinitializing filters!
Failed to inject frame into filter network: Invalid argument
Error while processing the decoded data for stream #0:0
Conversion failed!

FFmpeg version :
ffmpeg version 4.4 Copyright (c) 2000-2021 the FFmpeg developers
built with gcc 9 (Ubuntu 9.3.0-11ubuntu0~18.04.1)
configuration:
libavutil 56. 70.100 / 56. 70.100
libavcodec 58.134.100 / 58.134.100
libavformat 58. 76.100 / 58. 76.100
libavdevice 58. 13.100 / 58. 13.100
libavfilter 7.110.100 / 7.110.100
libswscale 5. 9.100 / 5. 9.100
libswresample 3. 9.100 / 3. 9.100

OS: Ubuntu 18.04

Thanks.

Hi,

I have followed instructions to create a Docker image here and successfully built and run a container with the following command:

$ docker run -it --privileged --net=host -v ~/.Xauthority:/root/.Xauthority -v /tmp/.X11-unix:/tmp/.X11-unix:rw -e DISPLAY=unix$DISPLAY -e HTTP_PROXY=$HTTP_PROXY -e HTTPS_PROXY=$HTTPS_PROXY -e http_proxy=$http_proxy -e https_proxy=$https_proxy -v /dev/video2:/dev/video2 -v ~/ffva/FFmpeg-patch:/FFmpeg-patch -v ~/ffva/data/models/intel:/root/intel_models:ro -v ~/ffva/data/models/common:/root/common_models:ro -e MODELS_PATH=/root/common_models -v ~/ffva/data/video:/root/video-examples:rw -e VIDEO_EXAMPLES_DIR=/root/video-examples ffmpeg-video-analytics-package:latest

Happily, the face detection sample works like a charm after some reconstruction from the commands in the scripts provided:

$ ffmpeg -i /root/video-examples/news.mkv -vf "detect=model=/root/face-detection-0102.xml:device=CPU,classify=model=/root/mobilenet-ssd.xml:model_proc=/root/age-gender-recognition-retail-0013.json:device=CPU,ocv_overlay" -y /root/video-examples/faces.mp4

However, I have been going through the Advanced usage examples and I keep receiving a Create network failed! message:

I would expect that

$ ffmpeg -i /root/video-examples/news.mkv -vf "detect=model=/root/mobilenet-ssd.xml:device=CPU:nireq=8,ocv_overlay" -y /root/video-examples/objects.mp4

Would also work, however this is the output I keep getting:

ffmpeg version 4.2 Copyright (c) 2000-2019 the FFmpeg developers
built with gcc 7 (Ubuntu 7.5.0-3ubuntu1~18.04)
configuration: --prefix=/usr --extra-cflags=-I/opt/intel/openvino/deployment_tools/inference_engine/include --extra-ldflags=-L/opt/intel/openvino/deployment_tools/inference_engine/lib/intel64 --libdir=/usr/lib/x86_64-linux-gnu --extra-libs='-lpthread -lm' --enable-shared --enable-gpl --enable-libass --enable-libfreetype --enable-openssl --enable-nonfree --enable-libdrm --enable-libmfx --disable-doc --disable-htmlpages --disable-manpages --disable-podpages --disable-txtpages --enable-libfdk-aac --enable-libmp3lame --enable-libopus --enable-libvorbis --enable-libx264 --enable-libx265 --enable-libjson_c --enable-libsrt --enable-libinference_engine_c_api --enable-libopencv --enable-python3
libavutil 56. 31.100 / 56. 31.100
libavcodec 58. 54.100 / 58. 54.100
libavformat 58. 29.100 / 58. 29.100
libavdevice 58. 8.100 / 58. 8.100
libavfilter 7. 57.100 / 7. 57.100
libswscale 5. 5.100 / 5. 5.100
libswresample 3. 5.100 / 3. 5.100
libpostproc 55. 5.100 / 55. 5.100
Input #0, matroska,webm, from '/root/video-examples/news.mkv':
Metadata:
COMPATIBLE_BRANDS: iso6avc1mp41
MAJOR_BRAND : dash
MINOR_VERSION : 0
ENCODER : Lavf58.35.101
Duration: 00:10:18.98, start: -0.007000, bitrate: 709 kb/s
Stream #0:0: Video: h264 (Main), yuv420p(tv, bt709, progressive), 1280x720 [SAR 1:1 DAR 16:9], 25 fps, 25 tbr, 1k tbn, 50 tbc (default)
Metadata:
HANDLER_NAME : ISO Media file produced by Google Inc.
DURATION : 00:10:18.960000000
Stream #0:1(eng): Audio: opus, 48000 Hz, stereo, fltp (default)
Metadata:
DURATION : 00:10:18.981000000
Stream mapping:
Stream #0:0 -> #0:0 (h264 (native) -> h264 (libx264))
Stream #0:1 -> #0:1 (opus (native) -> aac (native))
Press [q] to stop, [?] for help
Loading model: device=CPU, path=/root/mobilenet-ssd.xml
Setting batch_size=1, nireq=8
libavfilter/inference_backend/openvino_image_inference.c:364 : OpenVINOImageInferenceCreate Create network failed!
Assertion ret == 0 failed at libavfilter/inference_backend/ff_inference_impl.c:284
Aborted (core dumped)

Any hints are much appreciated