Good IPCameras that support OpenIPC

[90total]

New to IPsec cam setups, setting up my own home IPcam network with my home assistant w/ Frigate & Coral and wanted to ask the people who know more on this the following:

What IPcameras have we all used with open source firmware that have overall good day/night performance.

Just trying to get an idea of options before I commit to anything.

Thanks for reading & your thoughts

Any data's good data.... right?

 

[TonyR]

I would advise a newb (at least to IPCT) to instead focus their attention on camera sensor size to resolution ratio, focal length, and form factor, among others.
By sticking with a name brand camera such as Dahua, Hikvision, Uniview and a few others OEM'd by Dahua and Hikvision the firmware will be optimized to perform the best with the aforementioned hardware features.

P.S. -Welcome to IPCT!

 

[90total]

I would advise a newb (at least to IPCT) to instead focus their attention on camera sensor size to resolution ratio, focal length, and form factor, among others.
By sticking with a name brand camera such as Dahua, Hikvision, Uniview and a few others OEM'd by Dahua and Hikvision the firmware will be optimized to perform the best with the aforementioned hardware features.

P.S. -Welcome to IPCT!

Thanks for the heads up & info, believe a more open ended question is in order so I edited the OP. I would go with OEM if it's open source as I'm planning on an almost entirely open source setup.... Because I don't cry enough

 

[wittaj]

Anything good in this arena camera wise is not open source.

Many people here have incorporated Dahua OEM cameras with Home Assistant and Frigate and Coral.

And then just isolate the cameras from the internet.

 

[smile4theIPC]

Having a parallel interest and aware that open source firmware would have to be flashed, I joined (just now) to "+1" the question.

It seems this topic would be worthy of its own subforum; the relevant info on current IPC models is hard to find and this place is among the first where many would look.

Flashing open source (or any third-party) firmware is not difficult as long as good documentation is available. The need could be filled here.

Who joins and immediately stirs up trouble?

 

[looney2ns]

I have not seen any open source cams mentioned.

 

[TonyR]

+1^^^.
Same here.

 

[smile4theIPC]

I have not seen any open source cams mentioned.

+1^^^.
Same here.

No argument on this point. I'm likewise unable to find even a single open source offering in the market. (And guessing OP had the same experience, hence the question.)

But countless people are running them by flashing firmware from the linked source above. IOW, one has to buy the cam, then make it open source by installing the new firmware.

Pretty sure supporting this would be complementary to this site's business model (the ad-free experience of which is much appreciated). One would expect that the number-of-units-sold metric would be most important to the supplier(s), and that they would care not-at-all about what happens after the sale.

 

[wittaj]

I am sure they exist, but the only open-source cameras I have seen or have seen people post about here and cheap stuff working with a Pi and as such are very limited in application as it is just the lens module and could never be placed outside in that condition, etc.

It seems more of a hobby than real world application.

For the basic operation of the camera (firmware that gets the best out of the sensor), I don't see why anyone thinks they could do better than the billion dollar companies that make them? If they are that good, then they should work for those companies.

This is different than say AI where people are creating open-source, third party AI to compliment that limited ability of the AI of a camera, but these are to compliment that basics of the camera.

 

[Keegan]

90total and smile4theIPC, which routes did you take? Asking as I'm also going in the same direction, but different application.

The rest of this post is just me explaining why open source is attractive to some of us. If you're not interested you can skip it.

Every IPC is just an IC and a lens until you give it a housing. There's nothing stopping anyone from giving a DIY camera a weather-proof shell for a few bucks. "Tube" + acrylic + silicone.
A "billion dollar company" is--by definition--a company that's very good at making money, not necessarily good at making a quality product. So someone can easily make a better product if they're focused on doing that. Chinese companies churning out product that hits the bare minimum for consumer grade while returning as much profit as possible to their stock holders are not those people, and aren't known for paying very well. An individual with the computer science background necessary to produce quality firmware code would likely be far better off not working for such a company. They'd be neither valued nor appreciated.

All products discussed here are made in China. They're not known for their quality; they're known for their quantity...and their obligation to do whatever their government tells them to.
What's the best Hikvision on Amazon right now? <searches> 8MP for $138. I can slap together a raspi and an IMX708 for $50, saving $90 and get 12MP. I've done it. It works great. Oh, right, add two more bucks for the servos and a few more for the shell. $15 if you want PoE. But we don't always need 12MP, and we don't always want to build everything ourselves, right? Just because I can make a nice camera doesn't mean I want to or have the time to, and I need a bunch! Option #2: These billion dollar companies have two things going for them: economy of scale, and proprietary firmware. Open source firmware exploits the first by eliminating the second. That's why we're interested in it. If we can buy some $30 "good enough" Chinesium camera, save time and eliminate the security vulnerabilities while dramatically expanding the software options, flexibility and real world applications of the product, that'd be great, wouldn't it?

 

[wittaj]

90total and smile4theIPC, which routes did you take? Asking as I'm also going in the same direction, but different application.

The rest of this post is just me explaining why open source is attractive to some of us. If you're not interested you can skip it.

Every IPC is just an IC and a lens until you give it a housing. There's nothing stopping anyone from giving a DIY camera a weather-proof shell for a few bucks. "Tube" + acrylic + silicone.
A "billion dollar company" is--by definition--a company that's very good at making money, not necessarily good at making a quality product. So someone can easily make a better product if they're focused on doing that. Chinese companies churning out product that hits the bare minimum for consumer grade while returning as much profit as possible to their stock holders are not those people, and aren't known for paying very well. An individual with the computer science background necessary to produce quality firmware code would likely be far better off not working for such a company. They'd be neither valued nor appreciated.

All products discussed here are made in China. They're not known for their quality; they're known for their quantity...and their obligation to do whatever their government tells them to.
What's the best Hikvision on Amazon right now? <searches> 8MP for $138. I can slap together a raspi and an IMX708 for $50, saving $90 and get 12MP. I've done it. It works great. Oh, right, add two more bucks for the servos and a few more for the shell. $15 if you want PoE. But we don't always need 12MP, and we don't always want to build everything ourselves, right? Just because I can make a nice camera doesn't mean I want to or have the time to, and I need a bunch! Option #2: These billion dollar companies have two things going for them: economy of scale, and proprietary firmware. Open source firmware exploits the first by eliminating the second. That's why we're interested in it. If we can buy some $30 "good enough" Chinesium camera, save time and eliminate the security vulnerabilities while dramatically expanding the software options, flexibility and real world applications of the product, that'd be great, wouldn't it?

Let's see some examples of your 12MP camera video at night under low light conditions...

Prove to us you can make a camera that performs better cheaper....

 

[TonyR]

90total and smile4theIPC, which routes did you take? Asking as I'm also going in the same direction, but different application.

The rest of this post is just me explaining why open source is attractive to some of us. If you're not interested you can skip it.

Every IPC is just an IC and a lens until you give it a housing. There's nothing stopping anyone from giving a DIY camera a weather-proof shell for a few bucks. "Tube" + acrylic + silicone.
A "billion dollar company" is--by definition--a company that's very good at making money, not necessarily good at making a quality product. So someone can easily make a better product if they're focused on doing that. Chinese companies churning out product that hits the bare minimum for consumer grade while returning as much profit as possible to their stock holders are not those people, and aren't known for paying very well. An individual with the computer science background necessary to produce quality firmware code would likely be far better off not working for such a company. They'd be neither valued nor appreciated.

All products discussed here are made in China. They're not known for their quality; they're known for their quantity...and their obligation to do whatever their government tells them to.
What's the best Hikvision on Amazon right now? <searches> 8MP for $138. I can slap together a raspi and an IMX708 for $50, saving $90 and get 12MP. I've done it. It works great. Oh, right, add two more bucks for the servos and a few more for the shell. $15 if you want PoE. But we don't always need 12MP, and we don't always want to build everything ourselves, right? Just because I can make a nice camera doesn't mean I want to or have the time to, and I need a bunch! Option #2: These billion dollar companies have two things going for them: economy of scale, and proprietary firmware. Open source firmware exploits the first by eliminating the second. That's why we're interested in it. If we can buy some $30 "good enough" Chinesium camera, save time and eliminate the security vulnerabilities while dramatically expanding the software options, flexibility and real world applications of the product, that'd be great, wouldn't it?

You've likely experienced a lot more IP camera technology in your lifetime than I have...if you're really 124 as your profile says.

 

[steve1225]

New to IPsec cam setups, setting up my own home IPcam network with my home assistant w/ Frigate & Coral and wanted to ask the people who know more on this the following:

I big fan of Frigate & Coral, but I must mention - AI detection done in Frigate (using defaults AI models) are bad.. lots of false detections..

Default AI IVS detections in Dahua 5xxx (Pro line) is levels above what Frigate at this day is offering. And this is done in camera, You can use this events in Home Assistants...

ps. Dahua cameras (especially Dahua / Loryta 5442 models) are recommended by Frigate in theirs documentation Recommended hardware | Frigate

 

[smile4theIPC]

90total and smile4theIPC, which routes did you take? Asking as I'm also going in the same direction, but different application...

Still in the planning stages so haven't chosen hardware or arrangement yet. (Love this stuff, just not enough time to work on now.)

Flashing the open source firmware is not optional for me. It will be done. The missing data needed for good decision-making is, what hardware is inside which IPC model with which features and costs. Someone may have links to such info; I've not been able to find it.

Related, but OT for this thread, is data storage arrangement. A smart crook will recognize decoys and uploading to the cloud is not an option here. Yagi antennas to the neighbor's with mutual off-site storage is most appealing so far. Hopefully can get to this by summer.

 

[Keegan]

(Love this stuff, just not enough time to work on now.)

<knowing/concerned nod w/ 100 yard stare>
I would need another 100 years to finish all the projects I've thought up that I want to work on, but if I lived another 100, I'd need another 1000.

Flashing the open source firmware is not optional for me. It will be done. The missing data needed for good decision-making is, what hardware is inside which IPC model with which features and costs. Someone may have links to such info; I've not been able to find it.

Related, but OT for this thread, is data storage arrangement. A smart crook will recognize decoys and uploading to the cloud is not an option here. Yagi antennas to the neighbor's with mutual off-site storage is most appealing so far. Hopefully can get to this by summer.

<squint> You're even more paranoid than I am. Bravo. Also an advanced degree infosec grad, broken like Batman in more ways than you wish, or both?
Joking aside, I concur. If I can't at the very least hack into the existing software (crap as most are) and make adjustments to ensure it's operating the way I want (securely) I don't want it. Flashing with crowd-sourced firmware is best, and even if it's not perfect, you can just make it better. Then you know what code is running in your hardware, with all the benefits that provides.
Regardless, I wish us both the best of luck, as I suspect we've very different applications and use-cases.

You've likely experienced a lot more IP camera technology in your lifetime than I have...if you're really 124 as your profile says.

This gave me pause, then a chuckle, so thank you. "124? 124 what?...oh, age! Ha!" Information security background. Need-to-know. Does this forum need to know my age? No. 01/01/1900 it is! <grin> Really, what's the point of putting your age in a profile? There's no verification process, so it's worthless, as my input demonstrates. On the Internet, no one knows you're really a dog...
Though if you were in 'nam, sir, our experience is likely equal, as the tech is younger than both of us. My pfp is older than I'd care to admit.

Let's see some examples of your 12MP camera video at night under low light conditions...

Prove to us you can make a camera that performs better cheaper....

That is your takeaway from my post?
"Oh yeah? Prove it!"?
That seems fun! Here, let me try:

Anything good in this arena camera wise is not open source.

Oh yeah? Prove it!
Boy, I sure showed you! Zero effort on my part, and now you have to waste effort supplying me with evidence or I win!

Seriously, remember when I prefaced much of my post with this?

The rest of this post is just me explaining why open source is attractive to some of us. If you're not interested you can skip it.

It seems you should have skipped it, because that clearly isn't what you're after here.
There are two ways this can go.
1. I waste even more of my time on this forum getting and providing what you ask for. You move the goalposts and demand I get to the new ones. Absolutely no one will enjoy this. This repeats until I either get tired of it, or haven't implemented one feature you deem vital, at which point you'll claim your opinion is the correct one, thereby refuting everything I've ever said for all time, all while having put forth very little effort. I may not be 124, but I've learned my time is better spent not giving cookies to mice on the Internet. (Though teaching them why I won't is moderately entertaining from time to time.)
2. I simply recognize that I've nothing to prove, zero incentive nor obligation to kowtow to demands, that you've no desire to learn, seemingly only wish to argue in defense of the opinion you're (oddly?) emotionally invested into, and so I instead get back to...literally anything else. In the end I've achieved my goal of providing some benefits and insight into the open source community for those interested, and you...I dunno. I get the impression you'll pat yourself on the back for believing you've won an argument, even if you were the only one participating in it. This seems like the best outcome for everyone!

To avoid any confusion: I've chosen option #2.

For everyone else reading this, do yourself a favor and don't get stuck in the trap of thinking "Big business" always means "Ultimate product". Before we were hacking IPCam firmware we were hacking god-awful router firmware and the improvements were immense there, too. You also got continued support and zero-day updates to exploits. Businesses operate on "Good enough" principles. Passion projects in the open source community can vastly exceed COTS products. I say "can" with emphasis because code maturity is a thing and literally anyone can start a new project for any hardware--this is why we get questions like the one in OP. It's absolutely not for everyone. Open Source is almost always lot more work (Hence 90total's statement "Because I don't cry enough") but just like doing work on your own car or house, if it's the right solution, you end up with a better job done because the job matters to you. Please remember that being closed-minded, dismissive, passive-aggressive (or outright aggressive) towards those inquiring about or dabbling in adjacent technologies will not only discourage them from remaining in your community, it will outright eliminate any chance of them sharing any of the benefits they encounter with you. When all you have is a hammer...maybe it's time to shop around for more tools.

Good luck to us all in our projects going forward. Cheers.
Edit: It seems it wasn't clear to all parties involved: the above statement means I've found this forum won't provide the insights I was hoping for and wish you all well as I'm unlikely to return.

 

[wittaj]

<knowing/concerned nod w/ 100 yard stare>
I would need another 100 years to finish all the projects I've thought up that I want to work on, but if I lived another 100, I'd need another 1000.

<squint> You're even more paranoid than I am. Bravo. Also an advanced degree infosec grad, broken like Batman in more ways than you wish, or both?
Joking aside, I concur. If I can't at the very least hack into the existing software (crap as most are) and make adjustments to ensure it's operating the way I want (securely) I don't want it. Flashing with crowd-sourced firmware is best, and even if it's not perfect, you can just make it better. Then you know what code is running in your hardware, with all the benefits that provides.
Regardless, I wish us both the best of luck, as I suspect we've very different applications and use-cases.


This gave me pause, then a chuckle, so thank you. "124? 124 what?...oh, age! Ha!" Information security background. Need-to-know. Does this forum need to know my age? No. 01/01/1900 it is! <grin> Really, what's the point of putting your age in a profile? There's no verification process, so it's worthless, as my input demonstrates. On the Internet, no one knows you're really a dog...
Though if you were in 'nam, sir, our experience is likely equal, as the tech is younger than both of us. My pfp is older than I'd care to admit.



That is your takeaway from my post?
"Oh yeah? Prove it!"?
That seems fun! Here, let me try:

Oh yeah? Prove it!
Boy, I sure showed you! Zero effort on my part, and now you have to waste effort supplying me with evidence or I win!

Seriously, remember when I prefaced much of my post with this?

It seems you should have skipped it, because that clearly isn't what you're after here.
There are two ways this can go.
1. I waste even more of my time on this forum getting and providing what you ask for. You move the goalposts and demand I get to the new ones. Absolutely no one will enjoy this. This repeats until I either get tired of it, or haven't implemented one feature you deem vital, at which point you'll claim your opinion is the correct one, thereby refuting everything I've ever said for all time, all while having put forth very little effort. I may not be 124, but I've learned my time is better spent not giving cookies to mice on the Internet. (Though teaching them why I won't is moderately entertaining from time to time.)
2. I simply recognize that I've nothing to prove, zero incentive nor obligation to kowtow to demands, that you've no desire to learn, seemingly only wish to argue in defense of the opinion you're (oddly?) emotionally invested into, and so I instead get back to...literally anything else. In the end I've achieved my goal of providing some benefits and insight into the open source community for those interested, and you...I dunno. I get the impression you'll pat yourself on the back for believing you've won an argument, even if you were the only one participating in it. This seems like the best outcome for everyone!

To avoid any confusion: I've chosen option #2.

For everyone else reading this, do yourself a favor and don't get stuck in the trap of thinking "Big business" always means "Ultimate product". Before we were hacking IPCam firmware we were hacking god-awful router firmware and the improvements were immense there, too. You also got continued support and zero-day updates to exploits. Businesses operate on "Good enough" principles. Passion projects in the open source community can vastly exceed COTS products. I say "can" with emphasis because code maturity is a thing and literally anyone can start a new project for any hardware--this is why we get questions like the one in OP. It's absolutely not for everyone. Open Source is almost always lot more work (Hence 90total's statement "Because I don't cry enough") but just like doing work on your own car or house, if it's the right solution, you end up with a better job done because the job matters to you. Please remember that being closed-minded, dismissive, passive-aggressive (or outright aggressive) towards those inquiring about or dabbling in adjacent technologies will not only discourage them from remaining in your community, it will outright eliminate any chance of them sharing any of the benefits they encounter with you. When all you have is a hammer...maybe it's time to shop around for more tools.

Good luck to us all in our projects going forward. Cheers.

Dude since you are questioning my reading comprehension, let me remind YOU what you posted, because I did read it and thus why I asked what I asked:

"What's the best Hikvision on Amazon right now? <searches> 8MP for $138. I can slap together a raspi and an IMX708 for $50, saving $90 and get 12MP. I've done it. It works great."

So if you have done it and it works great, then it should be easy to prove it with a sample video of low light motion....

Everyone's idea of "it works great" is relative....

You are the NOOB here and don't know me or my reputation here, nor most here. Many here love tinkering and would build our own IF someone can prove their statement. People here have written programs and code to get more out of these cameras, so it isn't like we are a bunch of close minded dismissive snobs that you are alluding to...but we do wanna see proof.

We would much rather not have firmware source code that can be calling home and who knows what, which is why we isolate the cameras so that they cannot talk to the internet and phone home. We are a bunch of paranoid folks here, but if we can't make our own that is better than what we can buy, well we buy it and put the appropriate steps in place to mitigate the danger of a closed source firmware.

As I mentioned in my post that you took a snippet out of and tried to throw "prove it" back to me, I also stated we have yet to see someone post proof they could make one that works well in low light with motion. I have played with pi and arducams and like I said, it seemed more hobby than real world application of what folks use cameras here for, so if you can prove to me I am wrong since you have already done it and it works great, I am all eyes on it.

I don't move the goalposts - I am simply asking for you to prove your statement...which I take it you can't and you would rather "move the goalposts" with a wall of text reply rhetoric than proving it

So the question now is can YOU back up your own statement or are you simply trolling

I mean after all you have done it and it works great, so it shouldn't take but a few seconds to upload a video example!

 

[wittaj]

This is usually the sample we get when someone says their camera "works great" and we ask for a sample image motion with at night with little light:



Yeah the person could tell it was them, but this is totally useless to the police for a stranger.

So prove to us that your homemade 12MP camera on the IMX708 sensor is a low light night time motion capturing machine....

 

[TonyR]

What's the best Hikvision on Amazon right now? <searches> 8MP for $138. I can slap together a raspi and an IMX708 for $50, saving $90 and get 12MP. I've done it. It works great.

I say "bullshit"......talk is cheap. Prove me wrong.
Post sample images; day time, at night, at night with movement.

 

[TonyR]

Let's see some examples of your 12MP camera video at night under low light conditions...

 

[forlotto]

Well dunno what is up with these answers open IPC cameras at the moment are based on chip being used... I wish it were easier to be honest I wish there was a list of cameras that were using specific chips I like you am looking for the same thing.

List of recommended SoCs - OpenIPC

openipc.org

These are the supported chips however you'll have to take apart your camera and figure out what chip it is using. All of the cheap cameras I have are currently unsupported by OPENIPC. I've taken them all apart to see the chips they are using.

So far no luck haven't purchased an OPENIPC camera yet unfortunate to say the least.

So many cams come with hard coded back doors these days kinda sucks that means that folks can just drop in and watch your stream with the hard coded password whenever they wish.

hankvision or hankvision_2016 or things like 111111 or 123456 are often used as hard coded passwords which makes your camera viewable by anyone who has this hard coded password.

OPENIPC aims to stop that I find this to be one of the if not the most valuable projects that needs our support in the camera space. I'm not afraid to hook up TTL and upgrade to an uncompromised firmware.