Multi Camera Solutions for Nvidia Jetson Nano

Things to know about using multiple cameras on Jetson Nano

The Nvidia Jetson series is known for its extraordinary performance in artificial intelligence with a small form factor. Since the Jetson TX1/TX2 could be too expensive to become popular among hobbyists, the lower-end cost-effective Jetson Nano was introduced in April 2019 for only $99, and it immediately becomes a huge competitor to another maker-favorite platform, the Raspberry Pi.

The Jetson Nano outperforms the RPi a lot, especially in camera applications where more computing power is needed, such as:

  • Drones
  • High-level robot vision
  • Self-driving cars
  • Neuron network
  • Movement tracking
  • Machine learning
  • Computer vision
  • Artificial intelligence

It’s even designed with the Raspberry Pi style 40 GPIO pins and MIPI CSI-2 camera connector to be compatible with existing accessories from the Raspberry Pi ecosystem, such as Pi camera modules and Pi HATs. As it’s designed like a Raspberry Pi, it’s hard to connect several cameras to the Jetson Nano board. However, that’s not a problem for Arducam, because we have already solved it on the Raspberry Pi platform, and those solutions can also be applied to the Jetson Nano.

3 multi camera solutions by Arducam


Arducam is dedicated to embedded and IoT camera solutions for single-board computers with our HD camera modules and camera adapters. We are advanced in multi-camera solutions on Raspberry Pi and can easily bring all those features to the Jetson Nano platform. Currently, we offer three methods to connect more than 1 camera to a single Jetson Nano, respectively multi-camera adapter board for sequential activating of 4 camera modules, stereo camera HAT for sync dual camera vision, and USB 3.0 camera shield for USB multi-camera solution.

Arducam multiplexer: the 4-in-1 adapter board for Jetson Nano


The multi-camera adapter board, also known as the camera multiplexer, can connect up to 4 MIPI CSI cameras to a single MIPI camera port on a Raspberry pi or Jetson Nano. The only issue is that it is TDM based, so only one camera can be activated at a time, and the adapter will switch between these cameras in a fast way with the help of software, and makes it like four cameras are working at the same time. It is suitable for the systems where simultaneous camera streaming is not required and only still image analysis is needed.

The USB3.0 Camera Shield


Another high-speed interface on Jetson Nano is the USB 3.0 port, which can offer 5Gbps bandwidth equivalent to 4GByte/s. Arducam has a USB 3.0 camera shield solution that can be also used on the Jetson Nano. You can connect multiple USB3 cameras to the Jetson Nano, as long as the total USB bandwidth is within the 4GByte/s limitation. Synchronization between the cameras requires the external trigger input features of the global shutter cameras.

The near future: stereo camera HAT


We just released a revolutionary stereo camera HAT for Raspberry pi, which allows you to connect two cameras working simultaneously to a single MIPI CSI-2 port on a Raspberry Pi. Most importantly, these two cameras are fully synchronized. Since the Jetson Nano uses the same camera connector as the RPi, it’s feasible for us to bring that to this platform. Arducam now is working to port this stereo camera HAT onto the Jetson Nano.

Jetson Nano + Arducam – Do more with multi cameras

The Jetson series is introduced to “automate everything” with its GPU power and AI performance, and they also released a project-ready DIY autonomous robot Kit called JetBot to showcase what the Jetson Nano is capable of. However, we think you can do much more with software from the community and multi-camera solutions from Arducam.

Let’s take the autonomous robot car as an example. Now you can use CUDA programming (a computing platform and API by Nvidia) and Robotic Operating System (ROS, where AI methods and algorithms are integrated) on the Jetson Nano. With our stereo camera HAT, you can add both front and rear view cameras to the Jetson Nano based robot or take advantages of the stereo vision, and it could be useful when you are in an industrywise project that requires simultaneous localization and mapping (SLAM) for better navigation or even augmented reality.

Wanna Use Synchronized Dual Camera for Raspberry Pi 4 or Stereo Camera HAT for Pi 3B+/3? Arducam Makes Them All Come True

Two camera Solution on pi 4, 3B+ and 3 – What you need to know about the background

Stereo Cameras on Raspberry Pi 4 3B+/3 - Synchronized Dual-Camera for Pi

When we talk about a new two-camera solution on RPi, we have to mention the background of the stereo Pi camera on this popular open source hardware platform. For a long time, advanced users of the Raspberry Pi community want to take video or images from multiple cameras simultaneously for 3D vids & pics, face/object detection, stereo robotic vision or advanced surveillance and livestreaming applications, so they want a real stereo camera for raspberry pi 3, 4, and other standard Pi boards. It was an unresolved problem limited by the RPi’s hardware design because the standard Raspberry pi models only have one camera port. Although Arducam has multi-camera adapter boards that let you connect up to 4 cameras to a single pi board, it only actives one camera at a time, and you have to switch between them.

The other alternative is to use network-based synchronization, but you still have to use one Raspberry pi board for each camera setup, and the capture process is still not at the exact same time which will introduce the movement or artifacts between multiple images.

New Stereoscopic Camera Breakout Board – Arducam’s Big Step Forward

Dual camera adapter board on Raspberry Pi 4

Now Arducam released a stereo camera HAT for Raspberry Pi which allows you to connect two 5MP OV5647 or two 8MP IMX219 Pi cameras to a single standard Pi board and takes images or video at the same time. More importantly, this binocular solution makes both camera fully synchronized. The highlight of this stereo camera HAT is that it incorporates Arducam proprietary ArduChip solution and extends the usage of the single pi camera port, and it cheats the Raspberry pi as if only one camera connected. It supports and is fully compatible with Raspistill commands for preview, and Raspivid command for video capture/processing with RPi’s default camera driver for its 5MP and 8MP cameras.

Panoramic (3D) photos and videos with Raspberry Pi 4

3D Video Cameras for Raspberry Pi 4B/3b+/3

With the latest Pi 4’s hardware, Arducam makes more camera applications possible for Raspberry Pi 4. This new HAT also supports other Arducam Raspberry Pi cameras and proprietary camera driver that works up to two 16MP, and as the Pi 4 is much more powerful, you can not only shoot two pictures at the same time. With the shared oscillator and I2C broadcasting, the two cameras are fully synchronized in nanosecond level. These features are excellent for a lot of applications, such as 3D Vision for IoT or panoramics photos and videos. Combined with a nice fisheye camera from Arducam and some image processing software, you can stitch a wonderful panorama with higher resolution and better accuracy.

Feature and Specification

  • Support all series Raspberry Pi boards, Pi HAT board style
  • Two MIPI camera input and one MIPI camera output
  • Support 1, 2 and 4lane MIPI cameras up-to 800Mbps/lane
  • Support Raspistill (preview) and Raspivid (processing) commands for 5MP/8MP pi cameras with half-resolution combine
  • Support Arducam RPI cameras up to 16MP with proprietary camera driver with full resolution combine
  • Support camera side-by-side combine (split – right left), channel 0 and channel 1 software switching on the fly
  • Support digital pan in half resolution combine, programmable pan speed control
  • Oscillator sharing requires extra cables, soldering skill needed or please use Arducam bundle kit
  • Typical Applications in Stereo Vision, 360 Camera and Advance Surveillance Camera
  • Contact us for other platform support

Real-time depth mapping – A Typical application in which this stereo solution can be used

So what is the point of two cameras? The short answer is you will get different information from each camera, and if it’s fully synchronized, it is real-time. The information difference then could be used in applications like distance and depth mapping. To put it simply, this is also a bionic solution because that’s like how eyes work. If you are a user of Google’s camera app, you might have used the “Len Blur” feature in it, which requires you to lift up the camera to take pictures, and that is also image information difference. In this feature, it calculates the distance to decide which object is in the front to stay clear and what else should be blurred in the background, so you can have a picture that seems like it’s from a DSLR lens. However, that is only a simple application used in our daily life, and you can have much more fancy projects like 3d modeling and AR/VR Cameras for Raspberry Pi 4.

We’ve already finished the prototype, and it will be available for bulk ordering soon!

For more details, please refer to this document page:

Introducing Arducam IoTai – The Ultimate IoT (Internet of Things) Board with Camera Support, Based on ESP32 and in the Shape of Arduino UNO


IoT Platform with ESP32

If you are interested in maker boards and IoT projects, you must have heard of ESP platforms such as the ESP-32 series. Considering the Wifi and Bluetooth features integrated, it is really an extremely cheap and cool gadget.

Cost-effective IoT Solution with Arduino

Arduino, on the other hand, is another widely used cost-effective open-source development platform for embedded and IoT applications. One thing we love about the Arduino platform is its IDE (Integrated Development Environment), which is extremely friendly to all users, including the new starters. Since both the Arduino and ESP32 are fantastic platforms, can we figure out a way to take advantages of them both?

Arducam IoTai – ESP32 based UNO board with camera support

Arducam ESP32 UNO PSRAM (Arducam IoTai) is an upgrade to its predecessor ESP32 UNO board. It’s a solution to IoT (Internet of Things) and simple AI (Artificial Intelligence) applications based on ESP-32. This board incorporates an external 4MByte PSRAM that enlarges the RAM space and makes image capture and processing possible. The camera interface now can directly connect most of the parallel port cameras without additional hardware, like OV2640, OV7670, OV7725, etc., and transfer the image wirelessly to be used.

You can read more details about this maker board on the B0192 product page.

The board is also designed for battery-powered portable devices with its onboard battery charging circuits and supports lithium battery power supply. The SD card slot is an extra add-on for those advanced applications that require larger storage.

Based on open-source face detection and recognition library, it supports face detection and recognition in low frame rate. Its low power feature makes it an ideal solution for home security and IoT camera applications. The Arducam team also provides customized design and manufacture services, contact us for more information.

IoT Camera Module – Multi Camera Array 

The Arducam board supports both Arduino IDE and ESP-IDF software development tools, keeps the same form of factors and pinout as the standard Arduino UNO R3 board, and makes the Arduino shields reusable. That also applies to our SPI camera multiplexer shield. You can connect our multi-camera adapter to the Arducam IoTai as you would expect on a UNO board.

Low-Cost Smart Edge AI Camera

The ESP32 processor has dual-core 32bit CPU which can run up to 240MHz with a processing power of 600 DMIPS. It is fast enough to handle some image processing tasks like object detection and tracking, or even AI neuron network recognition like face detection and recognition. We have a video demo where you can see the Arducam IoTai board is running neuron network facial recognition algorithm and sending the result to the web browser through WIFI connection.

The better-than-most IoT board for machine vision

The ESP32 is already a better platform than most others in terms of IoT applications, and we believe that this ESP32 based board with camera module is also capable of becoming a better IoT board for machine vision. There are mainly two reasons for this, and they are its easy-to-use compatibility and battery-related features.

One problem that arises during a project is the compatibilities issues when you plan to upgrade the hardware to newer and more powerful hardware. You will always want to replace as little current setup as possible. Since this Arducam IoTai has the same pinouts and form factor as the classic Arduino Uno R3 board, you can easily move to our board while keeping your current hardware setting, and the hardware connection should also be the same, whether it’s an Arduino shield or just jumper wire connections.

The second reason is the battery-related features. The board comes with a 2-pin lithium battery connector and onboard battery charging circuits, which are really helpful for embedded projects without extra power supplies for simplicity and integrity.

Build your own facial recognition/detection system within 5 minutes

With this board, you will be able to build a facial recognition/detection system very quickly. Just connect Arducam IoTai to your computer with a USB cable for power, and then the board will set up a hotspot called “Arducam_esp32s_camera.” Connect your computer to that hotspot WIFI, then you’ll be ready to interact with this board.

Type in your browser and hit Enter, you will be able to view the control panel of the camera features. You can toggle all kinds of settings of the OV2640 camera, like resolution, quality, brightness, contrast, AWB, AE, and so on. You can also toggle the “face recognition” feature on to see faced squared in the video feed.

Just click “Start Streaming” and you will soon the video feed on the right side of the user interface. Toggle the features mentioned above if you want an optimal image quality. Click the “Enroll face” button, and it will learn the face and shows “Hello.” If another face shows up in the camera, you will see it squared in red with an “intruder alert” notification

Live streaming with 4 cameras

As we’ve mentioned, this camera comes with the same pinouts and form factor as the Arduino UNO R3, so you can still have access to the Arduino shields, so you can use Arducam SPI cameras multiplexer shield and Arducam SPI cameras to see a video feed of 4 cameras on this board. This board comes unsoldered, so you need to solder it on your own if you plan to use the shields.

Another preparation needed is to install the 4 SPI cameras onto the multi-camera adapter board as the following picture shows, and then mount the adapter-camera combination to the board like you would do on an Arduino board.

Upload the code to the board with the Arduino IDE:

Then you will be able to see the 4-camera demo in the Arducam Host:

A robot car from UCTRONICS

On releasing this board, we introduced it to many of our distributors all around the world. Among them, our top-level distributor UCTRONICS is very interested in it and even starts a robot car project that is built around this board. They’ve built a lot of smart robot cars based either on the Arduino or the Raspberry Pi platform, and one thing that bothers them a lot it how to balance between the camera feature and the cost of the robot.

For UCTRONICS, a robot without the video feature will be less interesting. With the Arduino platform, they will go with a USB camera and a WIFI adapter to enable the Arduino robot to transfer images and video feeds wirelessly, which costs a lot and brings the price of the whole robot up. The Raspberry Pi robot, on the other hand, costs even much higher because you will need a relatively newer Raspberry Pi motherboard with wireless features and a Raspberry Pi camera, which add up to almost $50.

Our ESP32 PSRAM UNO board almost solves all their concerns. It’s a fully-featured IoT platform, and the built-in WIFI and camera module of Arducam IoTai make wireless video streaming easy. What’s more, this inexpensive IoT board has the same pinouts and form factor as the Arduino UNO R3, which is the very board they have been using for their previous robots, and that makes it easier for them to move all the previous works to the new robot.

We’ve already seen their hardware prototypes, and this new little robot supports line following, object following, obstacle avoidance, image capturing and video streaming to the mobile app, and some other buzzer, color, and light related features, and it’s programmable with scratch-based drag-and-drop visual programming software.

Camera Multiplexer for Raspberry Pi 4: Arducam Solutions for Your Applications, and the Coming Surprises We Prepare for You


Introduction – What stops you from using dual cameras on the Raspberry Pi?

Many of the Raspberry Pi users may notice that there is only one camera slot on the board. Actually, the processors on the latest Raspberry Pi models, like the Pi 3 and Pi 4, offer support for two cameras, but only one of them is brought out on standard Pi boards. Although the Raspberry Pi Compute Modules bring out both to use, they still need customized carrier boards and are not widely used in the maker space.

Problem – Why would you need multi-cam or stereo camera for raspberry pi 4 (and more)

What if you need multiple cameras for a special application? Well, currently the easiest solution might be – you combine multiple Raspberry Pi boards and cameras. It’s feasible, but that really increases the complexity of wiring, the consumption of power and the occupation of space. In the end, the cost of deployment might far exceed your expectations.

Solution – Arducam want to be your savior in multi-camera applications

In the last several years, Arducam has designed and released multi-camera adapters for Raspberry Pi in 2 kinds: one allows you to connect 2 Pi cameras to a single Pi Zero, and the other allows for 4 Pi cameras to be connected to one Pi in Model A&B series. Both camera multiplexers are compatible with the standard 5MP OV5647 and 8MP IMX219 Pi camera drivers, so feel free to connect your existing cameras to our multiplexer and see them work as they should be. That being said, there are actually no restrictions on which model you are plugging those adapters into. You can put the stereo adapter where a 4-cam multiplexer is in and vice-versa.

The key point and fundamental theory of these multi-camera adapter boards is multiplexing. The Raspberry Pi only allows one camera to be connected and only recognizes one connected camera, so switching between different channels will allow different camera video streams to be transmitted to the RPi’s onboard CSI port one at a time. Therefore, the Raspberry Pi still connects to a single camera, but you will be able to see in the perspective of multiple ones.

Application – 3D Scanner, Dual-Cam, Surveillance, Drone Cam and More

Use Case 1: 3D Scanner with Raspi (Raspberry Pi) Camera

Recreate the 3d model from multiple images taken from different angles.

Use Case 2: Dual-Camera (360 view) for Raspberry Pi 4/3B+/3A+/3

Taking multiple images from different directions and stitching them together.

*We’ve also had an SPI multi-camera application here:

Use Case 3: Enhanced Surveillance system with raspberry pi 4

Raspberry Pi-based surveillance camera, connecting multiple cameras to a single Pi, and pointing them at different locations to cover all areas.

Use Case 4: Raspi (Raspberry Pi) Camera for Drones

Most of the drone camera use FPV mode, but multi-camera setup will let you switch to side view or back view, so it’s even easier when running a mission.

Limitations – Asynchronous Switching

There are also many limitations with this solution: you can’t capture the image or video simultaneously. You can only switch to a camera and then shoot an image or video at a time. Because the cameras are running asynchronously, it might be at half of the frame rates during switching, and you have to drop the bad frames before getting the following good ones. The CSI-2 camera signals are very high-speed MIPI signals, which are subject to noise, attenuation, and interference. Therefore, extra-long cables and cascading is no longer supported.

Coming Next – True Stereo Camera, Two Pi Cameras Fully Synchronized

We will release a new true stereo camera very soon that let you connect two pi cameras at the same time and fully synchronized. Stay tuned with Arducam for more surprises.

Arducam for Raspberry Pi 4 HD Cameras Hand-on: Arducam Ensures Your Compatibilities While Exploring New Opportunities


Last week, we finally received the package that contained the Raspberry Pi 4s we had ordered and started testing our camera modules on them. The Good news is that the Raspberry Pi 4 still keeps backward compatibility with the previous camera modules and sensors, so feel free to go with Arducam if you are seeking for new camera modules for your Raspberry Pi.

Here is an hand-on video about all kinds of Arducam camera modules and adapters:

Best Camera Modules for Pi 4 from Arducam

Arducam has been in the Pi cameras market since 2013, and we have released a bunch of cameras for the Raspberry Pi platform. The following products are the representatives of our newest best-selling camera modules:

Arducam B0176 5MP Auto Focus Camera module

5MP Pi Cam with motorized lens compatible with Pi 4 and earlier models
See smarter in the same form factor, now with keyboard focus control
Autofocus Python script powered by OpenCV.

Arducam B0180 8MP Wide Angle IMX219 Drop-in Replacement

M12 lens 175 degrees diagonal wide-angle IMX219 lens-sensor combination
Drop-in replacement to fix perfectly in you V2 camera board
Optional Low distortion lens and autofocus lens replacement

Arducam B0174 16MP MIPI Camera Module

1/2.8−inch Sony IMX298 image sensor
4656 x 3496 Pixels, a programmable motorized focus lens
Only work with Arducam Drivers and examples

Arducam B0165 1MP Global Shutter MIPI Camera Module

1/4-inch OmniVision OV9281 global shutter monochrome image sensor
3 μm pixel size, M12 mount lens
Only work with Arducam Drivers and examples

Pan-tilt-zoom (PTZ) for raspberry pi 4

PTZ is an abbreviation for pan, tilt and zoom and reflects the movement options of the camera, and a pan–tilt–zoom camera (PTZ camera) is a camera that is capable of remote directional and zoom control. Arducam PTZ camera for the Raspberry Pi is real PTZ with servos for directional control and varifocal lens for optical zoom, and it supports autofocus and IR cut switch control.

Arducam B0167 Pan-Tilt-Zoom Camera Accessories

Varifocal lens with moveable lens and real optical zoom
Servo controller board for pan and tilt directional control
Work with Arducam 36mm camera board for Raspberry Pi

Multi-camera on raspberry pi 4

The upgrade V2.1 multi-camera adapter board is for both 5MP and 8MP pi cameras. The adapter board can connect 4 cameras maximum to a single pi board. It works in sequential and not simultaneously. The newest revision fixed the compatibility issue with Raspbian 9.9

Arducam B012001 Multi-Camera Adapter (4x)

Accommodate up to four 5 or 8MP Pi cameras on a multi-camera adapter board.
Work in sequential, not simultaneously
Support Raspberry Pi 4 and previous models

No InfraRed Filter (NoIR) cameras for raspberry pi 4

Sometimes you want a camera that is IR sensitive, then you need a lens without an IR filter. If you want both decent daylight picture and night vision, we have a camera module with program-controlled switchable IR filter and LED boards for day & night photography. We are also developing a newer model with a light sensor on the camera board to trigger the IR filter based on the lighting condition.

Arducam B003503 5MP Switchable IR Cut Filter Camera

Switchable IR cut filter control by software for both daylight and night vision
Light sensor-controlled IR LED boards and IR corrected lens
Work with Raspberry Pi 4 and earlier model. Include cable for Pi Zero

Raspberry Pi 4 and Arducam: New Opportunities and Challenges to Arducam Camera Modules


The Raspberry Pi 4 model B has been released on Jun 29th, and now you have access to up to 4GB of memory and 2 USB 3.0 ports, and that means a lot for Raspberry Pi 4 projects, especially for camera applications. With a boost in its performance, this long-anticipated Raspberry Pi sets a new standard for single board computers, and at the same time sets a series of opportunities and challenges for Arducam to face.

As we’re pretty sure that all the previous Raspberry Pi camera modules will continue to work with the Raspberry Pi 4, Arducam is still your reliable supplier to buy raspberry pi 4 cameras from. However, what makes us stand out is not just the basic camera modules, we’ve still got a lot of customized and exclusive camera modules and camera adapters different from the official camera modules that could be used with your Raspberry Pi 4.

The Raspberry Pi 4 offers more processing power for our high perfomance MIPI camera modules

Arducam is one of the first solution providers to offer high performance on Raspberry Pi and has released our MIPI camera product series for you to use global shutter cameras or high definition cameras directly through the MIPI CSI-2 port on the Raspberry Pi with the userland driver. That’s highly anticipated on Raspberry Pi, but the higher resolution requires more GPU memory and more compute performance, which makes it difficult for the previous Raspberry Pi models to handle. However, the Raspberry Pi 4 comes with an updated SoC and offers up to 4GB of memory so that we can explore more features and even higher resolutions on the Raspberry Pi 4.

With the two new USB 3 ports available on the Raspberry Pi, you will be able to use Arducam USB3 camera shield better. The Arducam USB camera shield is a general-purpose camera control board designed for both PCs and embedded systems like the Raspberry Pi. It supports almost all parallel interface image sensors ranging from 0.3MP to 16MP in global and rolling shutters. The Arducam USB3 camera shield comes with a comprehensive software SDK library and example source code you need to make it work on a Raspberry Pi 4. Fast camera evaluations on a Raspberry Pi just got faster and easier.

At the same time, we also need to make sure our current modules will continue to work with the Raspberry Pi 4. Many of our clients are concerned about whether our camera adapters could work with the Raspberry Pi 4. Arducam will make sure the Pi 4 multi-camera support continues with our multi camera adapters.