A client side implementation of the ONVIF specification.
libonvif is a multi platform library implementing the client side of the ONVIF specification for communicating with IP enabled compatible cameras. It will compile on Linux and Windows.
It has a comprehensive GUI sample program written in Python that includes the discovery functionality as well as controls for adjusting camera parameters and PTZ operations. The GUI program has a record function that will write the camera stream to file and includes some basic media file management tools. The GUI also has a module plug in function that allows developers to access the video stream in numpy format for python processing.
A utility program is included with libonvif that can be used as a maintenance tool and will discover compatible cameras on the local network and may be used to query each of them for device configuration such as RSTP connection uri information or video settings.
*Note: To install without Python and GUI interface, use CMAKE -DWITHOUT_PYTHON=ON ..
BUILD ON LINUX
The program has dependencies on FFMPEG, libxml2, libsdl2 and python. In order to run the onvif-gui program, you will need pip to install numpy, pyqt6 and opencv. If you are running an NVIDIA graphics card, you will need the proprietary NVIDIA drivers, the generic drivers are not stable with the GLWidget. Note that when cloning the project, the --recursive flag is needed for python bindings.
PREREQUISITES
sudo apt install git
sudo apt install g++
sudo apt install cmake
sudo apt install python3-pipBUILD
sudo apt install libxml2-dev
sudo apt install libavcodec-dev
sudo apt install libavdevice-dev
sudo apt install libsdl2-dev
sudo apt install python3-dev
git clone --recursive https://github.com/sr99622/libonvif.git
cd libonvif
mkdir build
cd build
cmake ..
make
sudo make install
sudo ldconfigBUILD ON WINDOWS
The recommended method for building libonvif on Windows is to use a conda environment to install dependencies. To install anaconda on Windows, please refer to the link https://docs.anaconda.com/anaconda/install/windows/. Once anaconda has been installed, launch a conda prompt and then use the following commands to build. You will need to have Microsoft Visual Studio installed with the C++ compiler, as well as git and cmake. The cmake installer will integrate the executables and development files into the conda environment. The conda environment must be active when running the executables.
conda create --name onvif -c conda-forge libxml2 ffmpeg sdl2 python
conda activate onvif
git clone --recursive https://github.com/sr99622/libonvif.git
cd libonvif
mkdir build
cd build
cmake -DCMAKE_INSTALL_PREFIX=%CONDA_PREFIX%\Library ..
cmake --build . --config Release
cmake --install .onvif-gui has a facility for incorporating python programs to operate on the video stream. The Modules tab is the user interface for this feature. There is a minimal example program called sample.py that demonstrates how data is trsansferred from the main program to the python module and it's GUI interface implementation.
There is included with onvif-gui a full implementation of the YOLOX algorithm along with an associated tracking algorithm known as ByteTrack. These algorithms are implemented using pytorch, which requires some specific configuration. Note that ByteTrack currently is available only on Windows.
BUILD ON LINUX WITH MODULES ENABLED
Ubuntu is strongly recommended if using linux.
Use the standard install instructions to install the onvif and avio libraries. Install pytorch and associated tools using pip. Follow the instructions above for install on linux, then continue with the instructions below.
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
pip install pyqt6 fvcore cloudpickle pycocotools fairscale timm omegaconf opencv-python scipy lap loguru
cd libonvif/onvif-gui
python3 main.pyBUILD ON WINDOWS WITH MODULES ENABLED
Conda is recommended for installing the dependencies required for running these algorithms. Note that pytorch has limited compatibility with respect to python versions as well as cuda versions. Please check your cuda version using nvidia-smi prior to installing pytorch. The instructions below assume a cuda version 11.7. Python 3.9 is the preferred version for this installation.
The full installation uses a combination of conda and pip to install, and the order of installation should be followed. After the dependencies have been installed, cython_bbox should be installed from the project source tree. This is an optimized installation of the bbox algorithm which depends on configuration so is run last.
conda create --name myenv -c conda-forge python=3.9 numpy ffmpeg sdl2_ttf libxml2 opencv scipy lap loguru cython
conda activate myenv
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
pip install pyqt6 fvcore cloudpickle pycocotools fairscale timm omegaconf
git clone --recursive https://github.com/sr99622/libonvif.git
cd libonvif/cython_bbox
python setup.py install
cd ..
mkdir build
cd build
cmake ..
make
cd ../onvif-gui
python main.pyNOTES ON RUNNING PYTORCH MODULES
When running these algorithms, bear in mind that they are computationally expensive. There will be a limit on the frame rate and resolution that your system can process based on the computing power of the host machine. Limiting the frame rate can be done on cameras using the video settings tab. Frame rate on files can be set by using the filter command 'fps=10' where 10 is the desired frame rate. There is currently a limitation in onvif gui that prevents seeking when the frame rate is set by the filter. Likewise, resolution can be set on files with the video filter using the scale directive, i.e. 'scale=1280x720'. Consecutive video filters can be run using a comma as delimiter between the commands, i.e. 'fps=10,scale=1280x720'.
NAME
onvif-gui
SYNOPSIS
onvif-gui is a python program.
The program requires the following modules:
pip install pyqt6 opencv-python numpy loguru
To run the program:
cd ../onvif-gui
python3 main.py
These instructions are intended for quick setup to verify the program. To use the
library in other python programs, it is advised to install the onvif and avio
python modules.
cd ../libonvif
python3 setup.py install
cd ../libavio
python3 setup.py install
DESCRIPTION
GUI program to view and set parameters on onvif compatible IP cameras. Double clicking
the camera name in the list will display the camera video output.
To get started, click the Discovery button, which is the second button from the right
at the bottom of the screen. A login screen will appear for each camera as it is found.
The Settings tab may be used to set a default login that can be used automatically.
Camera parameters are available on the tabs on the lower right side of the application.
Once a parameter has been changed, the Apply button will be enabled, which can be used
to commit the change to the camera. It may be necessary to re-start the video output
stream in order to see the changes. The Apply button is found in the lower right hand
corner below the tabs.
Video:
Resolution - The drop down box is used to select the setting.
Frame Rate - The number of frames per second in the video output.
Gov Length - This is the distance between key frames in the stream.
Bitrate - The maxmimum number of bits per second to transmit.
Image:
All values are set using the sliders
Brightness
Saturation
Contrast
Sharpness
Network:
If the DHCP is enabled, all fields are set by the server, if DHCP is disabled, other
network settings may be completed manually. Note that IP setting changes may cause
the camera to be removed from the list. Use the Discover button to find the camera.
Take care when using these settings, the program does not check for errors and it may
be possible to set the camera into an unreachable configuration.
IP Address
Subnet Mask
Gateway
Primary DNS
PTZ:
Settings pertain to preset selections or current camera position. The arrow keys, Zoom In
and Zoom out control the position and zoom. The numbered buttons on the left correspond to
preset positions. The blank text box may be used to address presets numbered higher than 5.
To set a preset, position the camera, then check Set Preset, then click the numbered preset button.
Admin:
Camera Name - Changes the application display name of the camera.
Set admin Password - Can be used to change the password for the camera.
Sync Time - Will reset the camera's current time without regard to time zone.
Browser - Will launch a browser session with the camera for advanced maintenance.
Enable Reboot - Will enable the reboot button for use. Camera will be removed from list.
Enable Reset - Will enable the reset button for use. Use with caution, all camera
settings will be reset.
Application Settings:
Auto Discovery - When checked, the application will automatcally start discovery upon launch,
otherwise use the Discover button.
Common Username - Default username used during discover.
Common Password - Default password used during discover.
Hardware Decoder - If available, can be set to use GPU video decoding.
Video Filter - FFMPEG filter strings may be used to modify the video
Direct Rendering - May be used in Windows to increase performance
Convert to RGB - The default setting is ON, may be turned on for performance
Disable Audio, Disable Video - Used to limit streams to a single medium
Post Process Record - Recording will be the encoded video stream rather than raw packets
Hardware Encode - If available, use the GPU for encoding
Process Frame - Video frame data is processed by the sample python module
Low Latency - Reduces the buffer size to reduce latency, may cause instability
Pre-Record Cache Size - A cache of media packets is stored locally prior to decoding and will
be pre-pended to the file stream when Pre Process recording. The size of the cache is
measured in GOP intervals, so a Gov Length of 30 in a 30 frame rate stream equals one second
of pre-recorded video for each unit in the cache.
Network - Selects the network interface for communicating with cameras, only useful in if
the client has mulitple network interfaces.
EXAMPLES
To change the video resolution of a camera output, Double click on the camera name in
the list. The camera video output should display in the viewer. Select the Video tab
and use the drop down box labelled Resolution. Upon changing the selection, the Apply
button will be enabled. Click the apply button to make the change. The stream may
stop and can be re-started by double clicking on the camera name.
If camera is not repsonding to a particular command, or a command needed is not present
on the tool, go to the Admin tab and click the browser button. This will launch the
browser using the camera IP address. Log into the camera and settings should be
avialable in native format for the camera configuration.
SEE ALSO
There is a command line version of this program included with the libonvif package which
will implement most of the same commands. It may be invoked using the 'onvif-util' command.
NOTES
Camera compliance with the onvif standard is often incomplete and in some cases
incorrect. Success with the onvif-util may be limited in many cases. Cameras
made by Hikvision will have the greatest level of compatibility with onvif-util.
Cameras made by Dahua will have a close degree of compatability with some notable
exceptions regarding gateway and DNS settings. Time settings may not be reliable
in some cases. If the time is set without the zone flag, the time appearing in
the camera feed will be synced to the computer time. If the time zone flag is used,
the displayed time may be set to an offset from the computer time based on the
timezone setting of the camera.
If the camera DNS setting is properly onvif compliant, the IP address may be reliably
set. Some cameras may not respond to the DNS setting requested by onvif-gui due
to non compliance. Note that the camera may reboot automatically under some conditions
if the DNS setting is changed from off to on.
Video settings are reliable. The Admin Password setting is reliable, as well as the reboot
command. If there is an issue with a particular setting, it is recommended to connect to
the camera with a web browser, as most cameras will have a web interface that will allow you
to make the changes reliably. The gui version has a button on the Admin tab that will launch
the web browser with the camera ip address automatically.
SYNOPSIS
onvif-util [-ahs] [-u <user>] [-p <password>] [host_ip_address]
DESCRIPTION
View and set parameters on onvif compatible IP cameras. The command may be used to
find and identify cameras, and then to create an interactive session that can be
used to query and set camera properties.
-a, --all
show all cameras on the network
-h, --help
show the help for this command
-u, --user
set the username for the camera login
-p, --password
set the password for the camera login
To view all cameras on the network:
onvif-util -a
To login to a particular camera:
onvif-util -u username -p password ip_address
To login to a camera with safe mode disabled:
onvif-util -s -u username -p password ip_address
Once logged into the camera you can view data using the 'get' command followed by
the data requested. The (n) indicates an optional profile index to apply the setting,
otherwise the current profile is used
Data Retrieval Commands (start with get)
get rtsp 'pass'(optional) (n) - Get rtsp uri for camera, with optional password credential
get capabilities
get time
get profiles
get profile (n)
get video (n)
get video options (n)
get imaging
get imaging options
get network
Parameter Setting Commands (start with set)
set resolution (n) - Resolution setting in the format widthxheight, must match option
set framerate (n)
set gov_length (n)
set bitrate (n)
set bightness value(required)
set contrast value(required)
set saturation value(required)
set sharpness value(required)
set ip_address value(required)
set default_gateway value(required)
set dns value(required)
set dhcp value(required) - Accepted settings are 'on' and off'
set password value(required)
Maintenance Commands
help
safe - set safe mode on. Viewer and browser are disabled
unsafe - set safe mode off. Viewer and browser are enabled
browser - Use browser to access camera configurations
view (n) - View the camera output using ffplay (ffplay must be installed in the path)
view player (n) - View the camera output with user specified player e.g. view vlc
sync_time 'zone'(optional) - Sync the camera time to the computer
dump - Full set of raw data from camera configuration
reboot
To Exit Camera Session
quit
EXAMPLES
A typical session would begin by finding the cameras on the network
> onvif-util -a
Looking for cameras on the network...
Found 8 cameras
192.168.1.18 localhost(TV TV-IP319PI)
192.168.1.7 (IPC-BO IPC-122)
192.168.1.14 IPC(Dahua IPC-HDW4631C-A)
192.168.1.6 IPC(Amcrest IP2M-841EB)
192.168.1.12 (AXIS M1065-LW)
192.168.1.12 (AXIS M1065-LW)
192.168.1.2 IPC(Amcrest IP3M-HX2W)
192.168.1.11 R2(IPC-model)
To start a session with a camera, use the login credentials
> onvif-util -u admin -p admin123 192.168.1.12
found host: 192.168.1.12
successfully connected to host
name: AXIS M1065-LW
serial: ACCC8E99C915
Get current settings for video
> get video
Profile set to profile_1_h264
Resolution: 1920 x 1080
Frame Rate: 25
Gov Length: 30
Bit Rate: 4096
Get available video settings
> get video options
Available Resolutions
1920 x 1080
1280 x 720
640 x 480
320 x 240
Min Gov Length: 1
Max Gov Length: 32767
Min Frame Rate: 1
Max Frame Rate: 30
Min Bit Rate: 1
Max Bit Rate: 2147483647
Set video resolution
> set resolution 1280x720
Resolution was set to 1280 x 720
Exit session
> quit
SEE ALSO
There is a GUI version of this program included with the libonvif package which will implement most of the same commands. It may be invoke using the 'onvif-gui' command.
Copyright (c) 2018, 2020, 2022, 2023 Stephen Rhodes
License: GPL-2+
This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
libavio Copyright (c) 2022, 2023 Stephen Rhodes
License: Apache
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
getopt-win.h (originally getopt.h) Copyright (c) 2002 Todd C. Miller Todd.Miller@courtesan.com and Copyright (c) 2000 The NetBSD Foundation, Inc.
License: BSD-2-Clause-NETBSD
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
- Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
- Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. . THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
cencode.h, cencode.c in Public Domain by Chris Venter : chris.venter[anti-spam]gmail.com
License: public-domain1
Copyright-Only Dedication (based on United States law) or Public Domain Certification
The person or persons who have associated work with this document (the "Dedicator" or "Certifier") hereby either (a) certifies that, to the best of his knowledge, the work of authorship identified is in the public domain of the country from which the work is published, or (b) hereby dedicates whatever copyright the dedicators holds in the work of authorship identified below (the "Work") to the public domain. A certifier, moreover, dedicates any copyright interest he may have in the associated work, and for these purposes, is described as a "dedicator" below.
A certifier has taken reasonable steps to verify the copyright status of this work. Certifier recognizes that his good faith efforts may not shield him from liability if in fact the work certified is not in the public domain.
Dedicator makes this dedication for the benefit of the public at large and to the detriment of the Dedicator's heirs and successors. Dedicator intends this dedication to be an overt act of relinquishment in perpetuity of all present and future rights under copyright law, whether vested or contingent, in the Work. Dedicator understands that such relinquishment of all rights includes the relinquishment of all rights to enforce (by lawsuit or otherwise) those copyrights in the Work.
Dedicator recognizes that, once placed in the public domain, the Work may be freely reproduced, distributed, transmitted, used, modified, built upon, or otherwise exploited by anyone for any purpose, commercial or non-commercial, and in any way, including by methods that have not yet been invented or conceived.
sha1.h, sha1.c in Public Domain by By Steve Reid steve@edmweb.com
License: public-domain2
100% Public Domain.
debian folder Copyright: 2022 Petter Reinholdtsen pere@debian.org
License: GPL-2+
This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
YOLOX Copyright (c) 2021-2022 Megvii Inc. All rights reserved.
License: Apache
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
If you use YOLOX in your research, please cite our work by using the following BibTeX entry:
@article{yolox2021,
title={YOLOX: Exceeding YOLO Series in 2021},
author={Ge, Zheng and Liu, Songtao and Wang, Feng and Li, Zeming and Sun, Jian},
journal={arXiv preprint arXiv:2107.08430},
year={2021}
}Without the guidance of Dr. Sun Jian, YOLOX would not have been released and open sourced to the community. The passing away of Dr. Sun Jian is a great loss to the Computer Vision field. We have added this section here to express our remembrance and condolences to our captain Dr. Sun. It is hoped that every AI practitioner in the world will stick to the concept of "continuous innovation to expand cognitive boundaries, and extraordinary technology to achieve product value" and move forward all the way.
ByteTrack
MIT License
Copyright (c) 2021 Yifu Zhang
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
cython_bbox
Faster R-CNN
The MIT License (MIT)
Copyright (c) 2015 Microsoft Corporation
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
THIRD-PARTY SOFTWARE NOTICES AND INFORMATION
This project, Faster R-CNN, incorporates material from the project(s) listed below (collectively, "Third Party Code"). Microsoft is not the original author of the Third Party Code. The original copyright notice and license under which Microsoft received such Third Party Code are set out below. This Third Party Code is licensed to you under their original license terms set forth below. Microsoft reserves all other rights not expressly granted, whether by implication, estoppel or otherwise.
- Caffe, (https://github.com/BVLC/caffe/)
COPYRIGHT
All contributions by the University of California: Copyright (c) 2014, 2015, The Regents of the University of California (Regents) All rights reserved.
All other contributions: Copyright (c) 2014, 2015, the respective contributors All rights reserved.
Caffe uses a shared copyright model: each contributor holds copyright over their contributions to Caffe. The project versioning records all such contribution and copyright details. If a contributor wants to further mark their specific copyright on a particular contribution, they should indicate their copyright solely in the commit message of the change when it is committed.
The BSD 2-Clause License
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
-
Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
-
Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
**END OF THIRD-PARTY SOFTWARE NOTICES AND INFORMATION
MIT License
Copyright (c) 2018 WANG Chenxi
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
detectron2
Detectron2 is released under the Apache 2.0 license.
Copyright (c) Facebook, Inc. and its affiliates.
Citing Detectron2
If you use Detectron2 in your research or wish to refer to the baseline results published in the Model Zoo, please use the following BibTeX entry.
@misc{wu2019detectron2,
author = {Yuxin Wu and Alexander Kirillov and Francisco Massa and
Wan-Yen Lo and Ross Girshick},
title = {Detectron2},
howpublished = {\url{https://github.com/facebookresearch/detectron2}},
year = {2019}
}