Arducam 16MP Autofocus Camera Module for Raspberry Pi with ABS Case, IMX519 16 Megapixel High Resolution Autofocus Camera for All Raspberry Pi Models

This is my second experience with Arducam's cameras. I previously aqcuired the HawkEye 64MP Camera during their crownfunding campaign, but at the time the camera came out, Rapsberry Pi was transitioning to using libcamera and Picamera2, so a combination of the constant updates and changes to the settings coupled with Arducam being a little behind in updating their drivers made it a bit of a headache to use. I was pleased to see that the process is now pretty straight forward and the custom libcamera-apps required to use this camera is right in line with RPi OS now, so the camera can function as a drop-in replacement in terms of code once you have this set up, which only takes a few minutes (same for the 64MP, FWIW). Very pleased with this! UPDATE - 12/1/25 - The current custom libcamera/rpicam dev and apps files needed to install the camera are current up to the latest Trixie if anyone is wondering. The process is pretty straightforward. You go to Arducam's website and find the page for your camera. Look for the 'wiki' link on the page. head down to the guide there and follow the steps to download and install the drivers, as well as to update config.txt with the overlay settings needed for your pi to see the camera. On reboot, the system should see it and you should be able to use the camera as usual, but be warned; unless you use (sudo) apt-mark hold on libcamera-dev and libcamera-app, when you update your system the drivers will be replaced with the ones from the repository and you will need to reinstall them to resume camera functionality. All that being said I'm extremely pleased with this camera. If you don't need the manual focus the official HQ offers, I would highly recommend this camera, which takes nice, clear photos and video. I'm intending to use mine with a customer Photogrammetry setup, which is very common as this camera is one of the popular choices for openscan (please forgive my literal inaugral test photogrammetry video of a sweet potato that look like it has a face on it)

Good camera - sharp images. The mounting bracket might be a plus for some people.

getting it to work was a bit tricky. Could use improvment on the documentation, especially setting boot config.txt

I wish I could give this less than 1 star. This camera is basically junk. 1) The "instructions" included in the box are literally just "plug cable into camera and raspberry pi." Okay, cool I guess. Lets follow the link to the instructions. First red flag: the page the instruction booklet included with the product links to a webpage that doesn't have the right product on it. Oh goody. What a great job arducam. Where is the correct webpage you ask? Oh you're on the right one, the big brains at the company just straight up don't know their own chinese crapware. The webpage has 2 different connection types listed. Neither one are for this camera. Okay, so I google around a bit and figure out how to connect the ribbon cable the right way. Now lets get onto the software side. Oh my god. What a giant disaster. They've, for some reason, decided to mix up the instructions for different pis and os all in one giant page. Jesus Christ. A preschooler could tell you how stupid of a design this is. Fine. Whatever. I can manage. I go jump through the hoops, download and run all the correct commands for my pi and OS versions. Get to the end. It. Doesn't. Work. Just straight up, no frills, doesn't work. Is there troubleshooting on the website? Nope. Does re-running the commands help? Nope. Does switching the camera port help? Nope. So either their own instructions don't work, or their product doesn't work. Do not give this company a dime. This is one of the most all around incompetent pieces of hardware and user design I have EVER has the immense displeasure of dealing with. The fact that I have to give them a single, solitary star hurts me. They deserve less than that. What an embarrassing, disgraceful disaster.

I made few tests but I see better quality than others like 5 or 8 MP. It only can be used with Bullseye+ OS. I used a 64bit OS in a RPi 3B+. At first I was able to use this camera but OV5647 neither IMX 219/8MP. I installed required drivers detailed in the user manual but after that any of them were detected. Also had problems with python Picamera2 module. This one was solved with the help of Arducam folks at their forum. The other problems were solved rapidly letting techs access the Pi remotely. Very good support team.

Bought two of these for testing. They worked out of the box with video to my stream but fuzzy. I altered the code to include the autofocus mode: gst_pipeline = ( "qtiqmmfsrc camera=0 af-mode=3 ! " "video/x-raw(memory:GBM),format=NV12,width=1920,height=1080,framerate=30/1 ! " "qtivtransform ! " "video/x-raw,format=BGR ! " "appsink drop=1 sync=0" ) # --- THE QUALCOMM PIPELINE WITH AUTOFOCUS --- # 1. qtiqmmfsrc: # - camera=0: Primary camera. # - af-mode=(0=Off, 1=Auto, 2=Macro, 3=Continuous Video, 4=Continuous Picture) The autofocus worked but these may be Macro modules as I can focus very good from 4" to about 10" then they won't focus on anything beyond 10" (Yes you can see focus working as you move near or far) Maybe I'm not doing something right. Anyway. Not happy.

I bought it for a open scanner project and it is very clear. I will get one more for my friend who ask me to build him one it's a good buy.

I wanted a good camera to use with motion, and thought I'd give this a try. Once it's been set up, it works great. The autofocus is crisp too! The included plastic housing is nice and provides protection to the board components and lens and a convenient way to mount. However, it requires the use of specially downloaded libcamera-dev drivers. So it does not work with Motion, which was my intended use. Probably have to wait until Motion officially supports it, or something else idk. So if you're only planning to use libcamera functionality, or the docs say it can be used with OctoEye or gstreamer or python, then this will probably be fine! Otherwise, find a different camera.