Easy to use Python camera interface for NVIDIA Jetson

Related tags

Deep Learningjetcam
Overview

JetCam

JetCam is an easy to use Python camera interface for NVIDIA Jetson.

  • Works with various USB and CSI cameras using Jetson's Accelerated GStreamer Plugins

  • Easily read images as numpy arrays with image = camera.read()

  • Set the camera to running = True to attach callbacks to new frames

JetCam makes it easy to prototype AI projects in Python, especially within the Jupyter Lab programming environment installed in JetCard.

If you find an issue, please let us know!

Setup

git clone https://github.com/NVIDIA-AI-IOT/jetcam
cd jetcam
sudo python3 setup.py install

JetCam is tested against a system configured with the JetCard setup. Different system configurations may require additional steps.

Usage

Below we show some usage examples. You can find more in the notebooks.

Create CSI camera

Call CSICamera to use a compatible CSI camera. capture_width, capture_height, and capture_fps will control the capture shape and rate that images are aquired. width and height control the final output shape of the image as returned by the read function.

from jetcam.csi_camera import CSICamera

camera = CSICamera(width=224, height=224, capture_width=1080, capture_height=720, capture_fps=30)

Create USB camera

Call USBCamera to use a compatbile USB camera. The same parameters as CSICamera apply, along with a parameter capture_device that indicates the device index. You can check the device index by calling ls /dev/video*.

from jetcam.usb_camera import USBCamera

camera = USBCamera(capture_device=1)

Read

Call read() to read the latest image as a numpy.ndarray of data type np.uint8 and shape (224, 224, 3). The color format is BGR8.

image = camera.read()

The read function also updates the camera's internal value attribute.

camera.read()
image = camera.value

Callback

You can also set the camera to running = True, which will spawn a thread that acquires images from the camera. These will update the camera's value attribute automatically. You can attach a callback to the value using the traitlets library. This will call the callback with the new camera value as well as the old camera value

camera.running = True

def callback(change):
    new_image = change['new']
    # do some processing...

camera.observe(callback, names='value')

Cameras

CSI Cameras

These cameras work with the CSICamera class. Try them out by following the example notebook.

Model Infared FOV Resolution Cost
Raspberry Pi Camera V2 62.2 3280x2464 $25
Raspberry Pi Camera V2 (NOIR) x 62.2 3280x2464 $31
Arducam IMX219 CS lens mount 3280x2464 $65
Arducam IMX219 M12 lens mount 3280x2464 $60
LI-IMX219-MIPI-FF-NANO 3280x2464 $29
WaveShare IMX219-77 77 3280x2464 $19
WaveShare IMX219-77IR x 77 3280x2464 $21
WaveShare IMX219-120 120 3280x2464 $20
WaveShare IMX219-160 160 3280x2464 $23
WaveShare IMX219-160IR x 160 3280x2464 $25
WaveShare IMX219-200 200 3280x2464 $27

USB Cameras

These cameras work with the USBCamera class. Try them out by following the example notebook.

Model Infared FOV Resolution Cost
Logitech C270 60 1280x720 $18

See also

  • JetBot - An educational AI robot based on NVIDIA Jetson Nano

  • JetRacer - An educational AI racecar using NVIDIA Jetson Nano

  • JetCard - An SD card image for web programming AI projects with NVIDIA Jetson Nano

  • torch2trt - An easy to use PyTorch to TensorRT converter

Comments
  • Camera works, Jetcam does not

    Camera works, Jetcam does not

    I am trying to get a Raspberry Pi v2 camera module working on a Jetson Xavier NX with Jetpack 4.4 installed.

    (Specifically, I want to use Jetcam because one of your other projects, https://github.com/NVIDIA-AI-IOT/trt_pose uses Jetcam in its live demo.)

    I know my camera is connected properly and working because if I run:

    gst-launch-1.0 nvarguscamerasrc ! nvoverlaysink
    

    ... I get a video image on screen immediately, no problem.

    However, running even the most basic example (csi_camera notebook), I always get errors:

    ---------------------------------------------------------------------------
    RuntimeError                              Traceback (most recent call last)
    /usr/local/lib/python3.6/dist-packages/jetcam-0.0.0-py3.6.egg/jetcam/csi_camera.py in __init__(self, *args, **kwargs)
         23             if not re:
    ---> 24                 raise RuntimeError('Could not read image from camera.')
         25         except:
    
    RuntimeError: Could not read image from camera.
    
    During handling of the above exception, another exception occurred:
    
    RuntimeError                              Traceback (most recent call last)
    <ipython-input-2-4d23bcae2fae> in <module>
          1 from jetcam.csi_camera import CSICamera
          2 
    ----> 3 camera = CSICamera(width=224, height=224, capture_width=1980, capture_height=1080, capture_fps=30)
    
    /usr/local/lib/python3.6/dist-packages/jetcam-0.0.0-py3.6.egg/jetcam/csi_camera.py in __init__(self, *args, **kwargs)
         25         except:
         26             raise RuntimeError(
    ---> 27                 'Could not initialize camera.  Please see error trace.')
         28 
         29         atexit.register(self.cap.release)
    
    RuntimeError: Could not initialize camera.  Please see error trace
    

    I've even tried the fix (hack?) suggested in https://github.com/NVIDIA-AI-IOT/jetcam/issues/12 but this makes no difference.

    Any advice on what to look for or what the issue could be?

    opened by anselanza 3
  • remove duplicate comma

    remove duplicate comma

    This duplicate comma causes an error on Jetpack 4.3 (OpenCV 4). error opening bin: could not parse caps "video/x-raw, , format=(string)BGR" Fix #17

    opened by borongyuan 3
  • camera can not initial

    camera can not initial

    Python 3.6.9 (default, Oct 8 2020, 12:12:24) [GCC 8.4.0] on linux Type "help", "copyright", "credits" or "license" for more information.

    from jetcam.csi_camera import CSICamera camera = CSICamera(width=224, height=224, capture_width=1080, capture_height=720, capture_fps=30) Traceback (most recent call last): File "/usr/local/lib/python3.6/dist-packages/jetcam-0.0.0-py3.6.egg/jetcam/csi_camera.py", line 24, in init RuntimeError: Could not read image from camera.

    During handling of the above exception, another exception occurred:

    Traceback (most recent call last): File "", line 1, in File "/usr/local/lib/python3.6/dist-packages/jetcam-0.0.0-py3.6.egg/jetcam/csi_camera.py", line 27, in init RuntimeError: Could not initialize camera. Please see error trace.

    opened by wangnan31415926 1
  • cv2.cpython-36m-aarch64-linux-gnu.so: undefined symbol

    cv2.cpython-36m-aarch64-linux-gnu.so: undefined symbol

    [email protected]:/usr/lib$ python3
    Python 3.6.8 (default, Jan 14 2019, 11:02:34)
    [GCC 8.0.1 20180414 (experimental) [trunk revision 259383]] on linux
    Type "help", "copyright", "credits" or "license" for more information.
    >>> from jetcam.usb_camera import USBCamera
    Traceback (most recent call last):
      File "<stdin>", line 1, in <module>
      File "/usr/local/lib/python3.6/dist-packages/jetcam-0.0.0-py3.6.egg/jetcam/usb_camera.py", line 3, in <module>
    ImportError: /usr/local/lib/python3.6/dist-packages/cv2.cpython-36m-aarch64-linux-gnu.so: undefined symbol: _ZTIN2cv3dnn14dnn4_v201809175LayerE
    >>>
    
    
    

    My HW is jetson nano and SW env is

    [email protected]:/usr/lib$ jetson-release
     - NVIDIA Jetson NANO/TX1
       * Jetpack 4.2 [L4T 32.1.0]
       * CUDA GPU architecture 5.3
     - Libraries:
       * CUDA 10.0.166
       * cuDNN 7.3.1.28-1+cuda10.0
       * TensorRT 5.0.6.3-1+cuda10.0
       * Visionworks 1.6.0.500n
       * OpenCV 4.0.1 compiled CUDA: YES
     - Jetson Performance: active
    [email protected]:/usr/lib$
    
    
    opened by hgnan 0
  • Install failure

    Install failure

    I runsudo python3 setup.py install

    I get the following:

    /usr/local/lib/python3.8/dist-packages/setuptools/command/install.py:34: SetuptoolsDeprecationWarning: setup.py install is deprecated. Use build and pip and other standards-based tools.
      warnings.warn(
    /usr/local/lib/python3.8/dist-packages/setuptools/command/easy_install.py:144: EasyInstallDeprecationWarning: easy_install command is deprecated. Use build and pip and other standards-based tools.
      warnings.warn(
    /usr/local/lib/python3.8/dist-packages/pkg_resources/__init__.py:123: PkgResourcesDeprecationWarning: 0.1.36ubuntu1 is an invalid version and will not be supported in a future release
      warnings.warn(
    /usr/local/lib/python3.8/dist-packages/pkg_resources/__init__.py:123: PkgResourcesDeprecationWarning: 0.23ubuntu1 is an invalid version and will not be supported in a future release
      warnings.warn(
    running bdist_egg
    running egg_info
    writing jetcam.egg-info/PKG-INFO
    writing dependency_links to jetcam.egg-info/dependency_links.txt
    writing top-level names to jetcam.egg-info/top_level.txt
    reading manifest file 'jetcam.egg-info/SOURCES.txt'
    adding license file 'LICENSE.md'
    writing manifest file 'jetcam.egg-info/SOURCES.txt'
    installing library code to build/bdist.linux-aarch64/egg
    running install_lib
    running build_py
    creating build/bdist.linux-aarch64/egg
    creating build/bdist.linux-aarch64/egg/jetcam
    copying build/lib/jetcam/csi_camera.py -> build/bdist.linux-aarch64/egg/jetcam
    copying build/lib/jetcam/__init__.py -> build/bdist.linux-aarch64/egg/jetcam
    copying build/lib/jetcam/usb_camera.py -> build/bdist.linux-aarch64/egg/jetcam
    copying build/lib/jetcam/camera.py -> build/bdist.linux-aarch64/egg/jetcam
    copying build/lib/jetcam/utils.py -> build/bdist.linux-aarch64/egg/jetcam
    byte-compiling build/bdist.linux-aarch64/egg/jetcam/csi_camera.py to csi_camera.cpython-38.pyc
    byte-compiling build/bdist.linux-aarch64/egg/jetcam/__init__.py to __init__.cpython-38.pyc
    byte-compiling build/bdist.linux-aarch64/egg/jetcam/usb_camera.py to usb_camera.cpython-38.pyc
    byte-compiling build/bdist.linux-aarch64/egg/jetcam/camera.py to camera.cpython-38.pyc
    byte-compiling build/bdist.linux-aarch64/egg/jetcam/utils.py to utils.cpython-38.pyc
    creating build/bdist.linux-aarch64/egg/EGG-INFO
    copying jetcam.egg-info/PKG-INFO -> build/bdist.linux-aarch64/egg/EGG-INFO
    copying jetcam.egg-info/SOURCES.txt -> build/bdist.linux-aarch64/egg/EGG-INFO
    copying jetcam.egg-info/dependency_links.txt -> build/bdist.linux-aarch64/egg/EGG-INFO
    copying jetcam.egg-info/top_level.txt -> build/bdist.linux-aarch64/egg/EGG-INFO
    zip_safe flag not set; analyzing archive contents...
    creating 'dist/jetcam-0.0.0-py3.8.egg' and adding 'build/bdist.linux-aarch64/egg' to it
    removing 'build/bdist.linux-aarch64/egg' (and everything under it)
    Processing jetcam-0.0.0-py3.8.egg
    Removing /usr/lib/python3.8/site-packages/jetcam-0.0.0-py3.8.egg
    Copying jetcam-0.0.0-py3.8.egg to /usr/lib/python3.8/site-packages
    jetcam 0.0.0 is already the active version in easy-install.pth
    
    Installed /usr/lib/python3.8/site-packages/jetcam-0.0.0-py3.8.egg
    Processing dependencies for jetcam==0.0.0
    Finished processing dependencies for jetcam==0.0.0
    

    import jetcam returns ModuleNotFoundError: No module named 'jetcam'

    What am I doing wrong?

    opened by master0v 1
  • Jetbot Camera Not Working- RuntimeError: Could not initialize camera.  Please see error trace.

    Jetbot Camera Not Working- RuntimeError: Could not initialize camera. Please see error trace.

    Hello, For some reason I can't get my camera to work again. For context, I tried to use a custom dataset from roboflow but then my kernel kept dying after installing roboflow. I reconfigured the right numpy and edited my .bashrc as I saw in NVIDIA's forum. But now the camera wont initialize. I know it works because it used to work before. I also am able to save a short video with it and able to call it in the terminal. But whenever I try to run a cell in Jupyter that requires the camera, it fails. I've tried restarting the camera too. But no luck :( Any help would be appreciated!

    opened by niiita 1
  • Camera ON LED continues to be on unless I restart the OS.

    Camera ON LED continues to be on unless I restart the OS.

    Hi,

    How can I close the camera after camera.unobserve(update_image, names='value') ? The camera ON LED continues to be on unless I restart the OS. I am using Logitech C270 USB camera. Is there a command to close the camera?

    opened by jam244 0
  • jetcam thread race - read thread and processing thread

    jetcam thread race - read thread and processing thread

    with camera.running = True, jetcam spawns a thread which reads into camera.value

    Now let's say we do, new_image = change['new'] and do some processing. I guess Python does shallow copying and only assigns a reference to the original image array in the new_image variable. So, effectively, new_image and camera.value are pointing to the same memory region. Lets say my processing-thread takes a very long time. In the mean time, camera.value is updated by jetcam-thread. This can cause a thread race. Is that right?

    opened by PhilipsKoshy 0
  • Cannot query video position: status=0, value=-1, duration=-1

    Cannot query video position: status=0, value=-1, duration=-1

    i tried camera = USBCamera(width=224, height=224, capture_width=640, capture_height=480, capture_device=0) the reply is [ WARN:0] global /home/nvidia/host/build_opencv/nv_opencv/modules/videoio/src/cap_gstreamer.cpp (933) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1

    opened by Chenhait 0
  • AttributeError: 'directional_link' object has no attribute 'link'

    AttributeError: 'directional_link' object has no attribute 'link'

    The beginning steps is OK. But when 'camera_link.link()' fail to execute and I got an error: `--------------------------------------------------------------------------- AttributeError Traceback (most recent call last) in ----> 1 camera_link.link()

    AttributeError: 'directional_link' object has no attribute 'link'`

    Don't know what is the reason.

    opened by watershade 0
Releases(v0.0.0)
Owner
NVIDIA AI IOT
NVIDIA AI IOT
A minimalist environment for decision-making in autonomous driving

highway-env A collection of environments for autonomous driving and tactical decision-making tasks An episode of one of the environments available in

Edouard Leurent 1.6k Jan 07, 2023
Python implementation of a live deep learning based age/gender/expression recognizer

TUT live age estimator Python implementation of a live deep learning based age/gender/smile/celebrity twin recognizer. All components use convolutiona

Heikki Huttunen 80 Nov 21, 2022
Human4D Dataset tools for processing and visualization

HUMAN4D: A Human-Centric Multimodal Dataset for Motions & Immersive Media HUMAN4D constitutes a large and multimodal 4D dataset that contains a variet

tofis 15 Nov 09, 2022
UmlsBERT: Clinical Domain Knowledge Augmentation of Contextual Embeddings Using the Unified Medical Language System Metathesaurus

UmlsBERT: Clinical Domain Knowledge Augmentation of Contextual Embeddings Using the Unified Medical Language System Metathesaurus General info This is

71 Oct 25, 2022
Pytorch cuda extension of grid_sample1d

Grid Sample 1d pytorch cuda extension of grid sample 1d. Since pytorch only supports grid sample 2d/3d, I extend the 1d version for efficiency. The fo

lyricpoem 24 Dec 03, 2022
API for RL algorithm design & testing of BCA (Building Control Agent) HVAC on EnergyPlus building energy simulator by wrapping their EMS Python API

RL - EmsPy (work In Progress...) The EmsPy Python package was made to facilitate Reinforcement Learning (RL) algorithm research for developing and tes

20 Jan 05, 2023
PyTorch implementation of our method for adversarial attacks and defenses in hyperspectral image classification.

Self-Attention Context Network for Hyperspectral Image Classification PyTorch implementation of our method for adversarial attacks and defenses in hyp

22 Dec 02, 2022
[NeurIPS 2021]: Are Transformers More Robust Than CNNs? (Pytorch implementation & checkpoints)

Are Transformers More Robust Than CNNs? Pytorch implementation for NeurIPS 2021 Paper: Are Transformers More Robust Than CNNs? Our implementation is b

Yutong Bai 145 Dec 01, 2022
Dark Finix: All in one hacking framework with almost 100 tools

Dark Finix - Hacking Framework. Dark Finix is a all in one hacking framework wit

Md. Nur habib 2 Feb 18, 2022
offical implement of our Lifelong Person Re-Identification via Adaptive Knowledge Accumulation in CVPR2021

LifelongReID Offical implementation of our Lifelong Person Re-Identification via Adaptive Knowledge Accumulation in CVPR2021 by Nan Pu, Wei Chen, Yu L

PeterPu 76 Dec 08, 2022
Exploration of some patients clinical variables.

Answer_ALS_clinical_data Exploration of some patients clinical variables. All the clinical / metadata data is available here: https://data.answerals.o

1 Jan 20, 2022
Denoising Diffusion Probabilistic Models

Denoising Diffusion Probabilistic Models This repo contains code for DDPM training. Based on Denoising Diffusion Probabilistic Models, Improved Denois

Alexander Markov 7 Dec 15, 2022
A criticism of a recent paper on buggy image downsampling methods in popular image processing and deep learning libraries.

A criticism of a recent paper on buggy image downsampling methods in popular image processing and deep learning libraries.

70 Jul 12, 2022
Art Project "Schrödinger's Game of Life"

Repo of the project "Team Creative Quantum AI: Schrödinger's Game of Life" Installation new conda env: conda create --name qcml python=3.8 conda activ

ℍ◮ℕℕ◭ℍ ℝ∈ᛔ∈ℝ 2 Sep 15, 2022
MAME is a multi-purpose emulation framework.

MAME's purpose is to preserve decades of software history. As electronic technology continues to rush forward, MAME prevents this important "vintage" software from being lost and forgotten.

Michael Murray 6 Oct 25, 2020
noisy labels; missing labels; semi-supervised learning; entropy; uncertainty; robustness and generalisation.

ProSelfLC: CVPR 2021 ProSelfLC: Progressive Self Label Correction for Training Robust Deep Neural Networks For any specific discussion or potential fu

amos_xwang 57 Dec 04, 2022
A PyTorch implementation of deep-learning-based registration

DiffuseMorph Implementation A PyTorch implementation of deep-learning-based registration. Requirements OS : Ubuntu / Windows Python 3.6 PyTorch 1.4.0

24 Jan 03, 2023
Orthogonal Jacobian Regularization for Unsupervised Disentanglement in Image Generation (ICCV 2021)

Orthogonal Jacobian Regularization for Unsupervised Disentanglement in Image Generation Home | PyTorch BigGAN Discovery | TensorFlow ProGAN Regulariza

Yuxiang Wei 54 Dec 30, 2022
JFB: Jacobian-Free Backpropagation for Implicit Models

JFB: Jacobian-Free Backpropagation for Implicit Models

Typal Research 28 Dec 11, 2022
Official Implementation of SimIPU: Simple 2D Image and 3D Point Cloud Unsupervised Pre-Training for Spatial-Aware Visual Representations

Official Implementation of SimIPU SimIPU: Simple 2D Image and 3D Point Cloud Unsupervised Pre-Training for Spatial-Aware Visual Representations Since

Zhyever 37 Dec 01, 2022