I'm not sure that I've seen enough demand for it over the years to warrant a product.
There isn't 100% agreement on 2.4Ghz vs 5Ghz, and don't even get started on cellular variants.
You still need power (so, a cable).
Your cameras are typically installed by professionals, some of whom are *against* wifi because it would make things even more DIY.
In a number of cases where a wireless camera *would* make sense there are other issues like distance that make an external directional antenna more practical, at which point you might as well just use a Ubiquti unit (and at $80 street price, I'm not sure a wifi option on a "real" camera would be any cheaper).
Then there are all the support headaches with wifi, congested networks (and the resultant video issues).
In short, I'm not sure anyone would be willing to pay what it would cost to sell a true 'professional' IP camera that didn't fall short of expectations.
Most wireless (802.11) cameras have fairly short range due to the embedded wireless modules; most have very low output power (Average 200mW/23dBm) @2.4Ghz; combined with an antenna ranging from 0db (Unit Gain) to 3dBi, the short range (~700' radius free space) is expected depending on the receiver sensitivity (assuming -65dBm). It is really up to the integrator to calculate the path requirements for wireless and spec out the equipment that is required. So, the short range should not be an issue in most residential applications, but again, like anything else, it has to be engineered based on the requirements.
Do we know that they haven't gotten any better?
The time might be right for the first, (AFAIK), IPVM Wireless Shootout.
Hik 'n Dah have both recently released several new models that look like regular Pro Gear.
A few years back there were quite a few more options for professional grade wireless IP cameras than there are today. Those models represented less than 10% of our sales but probably 50% of our support calls. This and the fact you still have to get power to the camera so POE is usually more convenient may be why there are so few offerings now.
My issue or concern with putting in wireless embedded cameras beyond all of the other really good reasons stated, is that when one part of the system fails (the wifi or the camera), both things are rendered useless. So you will need to stock extra spare cameras just in case the wifi radio goes bad. In addition if there are new camera models, capabilities, or features available you would have to wait for those to trickle down to the wireless embedded cameras, or possibly never get those features without having to put in a separate wireless radio and camera (back to square 1).
Also in many environments now (education, government) the last network you want to try to add a camera in is your wifi that is hosting 120 iPads/Chromebooks/BYOD devices on 2 APs. Even in the best RF environment, with QOS set up as best as possible, air-fairness is going to be rough on a camera stream.
802.xx is shared space - all devices are arguing and accommodating surrounding users. There are a lot of little dirty secrets about wireless access points switching to the least common denominator when differing connections are on the same ssid. Latency can go through the roof, image quality then sucks, customers complain. Leave wireless for non time sensitive stuff like web browsing and email.
If you think Hik had hacking issues before, just mix in cheap Wi Fi and see what real trouble is.
Here is a script that kills connected wireless cameras - It's illegal of course but for educational purposes, test this in house.....
Yup, used simply as am example of wireless insecurities. For P2P / PMP links use TDMA protocols like WORP, and other iterations.