Google Git
Sign in
android / platform / external / libxcam / 2354fd550437a6a017439ce15e46c1c2aa7ab4a1
commit2354fd550437a6a017439ce15e46c1c2aa7ab4a1[log] [tgz]
authorzongwave <[email protected]>Mon Apr 01 09:21:50 2019 +0800
committerZong Wei <[email protected]>Mon Apr 01 18:24:54 2019 +0800
tree4fa2fcde16e49d70d8fc58c3a6a7e42edb71821e
parentcf02947a61da589d5fca5f0d1677a0ab276a88d3 [diff]
dnn: abstract dnn inference engine class and derive object detection class

      * clone openvino inference engine
          git clone https://github.com/opencv/dldt.git

      * build openvino inference engien
          https://github.com/opencv/dldt/blob/2018/inference-engine/README.md

      * export openvino path to environment varibles
          $ export OPENVINO_IE_INC_PATH=${OPENVINO_INFERENCE_ENGINE_DIR}
          $ export OPENVINO_IE_LIBS_PATH=${OPENVINO_INFERENCE_ENGINE_DIR}/bin/intel64/Release/lib

        # NOTE: Fill in the ${OPENVINO_INFERENCE_ENGINE_DIR} - e.g. /home/usr/openvino/dldt/inference-engine/

      * test command
          $ ./test-dnn-inference --input cars_on_road.jpg --model-file vehicle-detection-adas-0002.xml
8 files changed
tree: 4fa2fcde16e49d70d8fc58c3a6a7e42edb71821e
  1. capi/
  2. doc/
  3. ext/
  4. m4/
  5. modules/
  6. pkgconfig/
  7. plugins/
  8. shaders/
  9. tests/
  10. tools/
  11. wrapper/
  12. xcore/
  13. .gitignore
  14. .gitmodules
  15. Android.mk
  16. AUTHORS
  17. autogen.sh
  18. ChangeLog
  19. code_style.sh
  20. configure.ac
  21. COPYING
  22. LICENSE
  23. Makefile.am
  24. METADATA
  25. NEWS
  26. NOTICE
  27. README.md
README.md

libXCam

Copyright (C) 2014-2019 Intel Corporation

libxcam core source code under the terms of Apache License, Version 2.0

Description:

libXCam is a project for extended camera features and focus on image quality improvement and video analysis. There are lots features supported in image pre-processing, image post-processing and smart analysis. This library makes GPU/CPU/ISP working together to improve image quality. OpenCL is used to improve performance in different platforms.

Features:

  • Image processing features
    • Advanced features
      • Automotive surround view(360) stitching (OpenCL/CPU/GLES)
        • Support bowl view 3D model stitching by 4 video input.
        • Enable geometry remap for WFoV camera calibration(intrinsic and extrinsic data).
        • Quality and performance improved (OpenCL/CPU/GLES).
        • CPU version upstreamed into AOSP for automotive surround view.
        • Enable Vulkan to improve performance.
      • 360 video stitching (Equirectangular mode via OpenCL)
        • Support 2-fisheye (>180 degree) video stream stitching.
        • Performance and quality improved.
      • DNN inference framework
        • Support pedestrian and vehicle detection.
      • Digital Video Stabilization
        • OpenCV feature-matched based video stabilization.
        • gyroscope 3-DoF (orientation) based video stabilization.
      • Blender: multi-band blender (OpenCL/CPU/GLES)
      • Noise reduction (OpenCL)
        • adaptive NR based on wavelet-haar and Bayersian shrinkage.
        • 3D-NR with inter-block and intra-block reference.
        • wavelet-hat NR (obsolete).
      • Wide dynamic range (WDR) (OpenCL)
        • histogram adjustment tone-mapping.
        • gaussian-based tone-mapping (obsolete).
      • Fog removal: retinex and dark channel prior algorithm (OpenCL)
        • dark channel prior algorithm based defog.
        • multi-scale retinex based defog (obsolete).
    • Basic pipeline from bayer to YUV/RGB format (OpenCL / AtomISP)
      • Gamma correction, MACC, color space, demosaicing, simple bilateral noise reduction, edge enhancement and temporal noise reduction.
    • 3A features
      • Auto whitebalance, auto exposure, auto focus, black level correction, color correction, 3a-statistics calculation.
  • Support 3rd party 3A lib which can be loaded dynamically
    • hybrid 3a plugin.
  • Support 3a analysis tuning framework for different features
  • Support smart analysis framework
    • Face detection interface/plugin.
  • Enable gstreamer plugin
    • xcamsrc, capture from usb/isp camera, process 3a/basic/advanced features.
    • xcamfilter, improve image quality by advanced features and smart analysis.

Prerequisite:

  • install gcc/g++, automake, autoconf, libtool, gawk, pkg-config
  • Linux kernel > 3.10
  • install ocl-icd-dev, ocl-icd-opencl-dev
  • If --enable-libcl, need compile ocl driver https://www.freedesktop.org/wiki/Software/Beignet/
  • If --enable-opencv, suggest opencv versions [3.0.0 - 3.4.3]http://opencv.org (or: https://github.com/opencv/opencv/wiki)
  • If --enable-gst, need install libgstreamer1.0-dev, libgstreamer-plugins-base1.0-dev
  • If --enable-aiq, need get ia_imaging lib which we don't support.
  • If --enable-render, need compile OpenSceneGraph library with configure option “-DOSG_WINDOWING_SYSTEM=X11” https://github.com/openscenegraph/OpenSceneGraph.git
  • If --enable-gles, need to install mesa-3d library
  • If --enable-vulkan, need to install mesa-3d library
  • If --enable-dnn, need to install OpenVino SDK

Building and installing:

  • Environment variable settings
    For different --prefix options, the environment variables may be different. Please set the environment variable according to the actual situation.
    --prefix=/usr/local:

    export LD_LIBRARY_PATH=/usr/local/lib/:$LD_LIBRARY_PATH
export GST_PLUGIN_PATH=/usr/local/lib/gstreamer-1.0:$GST_PLUGIN_PATH

    --prefix=/usr:

    export LD_LIBRARY_PATH=/usr/lib/:$LD_LIBRARY_PATH
export GST_PLUGIN_PATH=/usr/lib/gstreamer-1.0:$GST_PLUGIN_PATH

  • $ ./autogen.sh [options]

    --prefix=PREFIX         install architecture-independent files in PREFIX [default=/usr/local]
--enable-debug          enable debug, [default=no]
--enable-profiling      enable profiling, [default=no]
--enable-drm            enable drm buffer, [default=no]
--enable-aiq            enable Aiq 3A algorithm build, [default=no]
--enable-gst            enable gstreamer plugin build, [default=no]
--enable-libcl          enable libcl image processor, [default=yes]
--enable-opencv         enable opencv library, [default=no]
--enable-capi           enable libxcam-capi library, [default=no]
--enable-docs           build Doxygen documentation [default=no]
--enable-3alib          enable 3A lib build, [default=no]
--enable-smartlib       enable smart analysis lib build, [default=no]
--enable-gles           enable gles, [default=no]
--enable-vulkan         enable vulkan, [default=no]
--enable-render         enable 3D texture render, [default=no]
--enable-dnn            enable dnn inference, [default=no]

    For example:

    $ ./autogen.sh --prefix=/usr --enable-gst --enable-libcl --enable-opencv \
  --enable-smartlib --enable-profiling --enable-gles --enable-render --enable-dnn

  • $ make

  • $ sudo make install

Testing:

Reporting Bugs:

Mailing list

Maintainer:

Contributors: (orders by first name)