The DNNCam™ is useful in a wide range of applications – here are some of the most commonly asked questions people have about our device.
On-board processing dramatically reduces the device’s need for bandwidth, as well as cloud storage requirements. This makes the DNNCam a perfect fit for fog computing.
Real-world durability makes the DNNCam capable of surviving in environments where other devices fail.
There are other great features too – low power requirements, exceptional technical specs, small size – but on-board processing and durability are the game changers.
Because data transmission and storage have significant costs, particularly at scale.
A ‘dumb’ AI camera that transmits raw data for cloud storage and processing requires high speed network access, remote data storage, and cloud computing. If you have one or two of these ‘dumb’ AI cameras that you need to accommodate, these costs are somewhat marginal.
But if you have a few dozen cameras collecting data for storage and processing, the costs become substantial. Imagine a big box retailer with 500 locations, each with half a dozen cameras generating terabytes of raw data each day. The data processing and storage costs become tremendous.
DNNCam™ is the solution to the data transmission and storage cost problem. By leveraging the onboard NVIDIA GPU, the DNNCam can process the footage in realtime and transmit the output of the installed AI software.
Because, as a wise man once said, “No battle plan survives contact with the enemy.” You can’t plan for all the possibilities your camera system may encounter.
A camera designed to perform in a dry, climate controlled, and relatively dust free setting is going to fail in the real world at some point. When it does, the cost of repairing or replacing that failed camera is going to effectively double or triple the camera’s cost.
After all, the cost of a failed camera isn’t just a replacement. Maintenance and service personnel need to be dispatched to verify the camera has failed and install a replacement. Then, a new camera will need to be tested, and data verified. Etc.
With DNNCam™, unexpected changes in the environment aren’t a problem.
While it’s true that the DNNCam doesn’t have the processing power of a cloud-based system, we’d argue that you don’t need a server’s processing power for most AI camera applications. The NVIDIA TX2 GPU is incredibly robust, and works great for most contemporary AI camera applications.
With the continued development of computer technology, deep neural net processing of camera data is becoming easier and easier. While every application is different, we have yet to find a situation where DNNCam lacks sufficient processing power.
The DNNCam™ is designed to be compatible with most popular AI software and programming languages. The camera has the NVIDIA ‘Jetson’ TX2 GPU onboard, which works with most AI software.
Additionally, our company can facilitate loading or adapting existing software to our camera system.
Depending on your needs and application, we can provide a price for the camera as a stand-alone piece of hardware, OR we can provide a price for ongoing use of the camera that includes service and maintenance.
Many of our clients prefer that we provide the DNNCam™ as part of a larger service, as they lack AI systems expertise and/or bandwidth.
For a specific price quote, please contact us.