# What fps can be considered as a hardline for real-time performance? Is there any academic paper that describes this?

I am trying to design a deep learning based inferencing solution for security applications. My deployed program achieves an FPS of 15fps for classification. Can it be considered real-time?

• i think the wikipedia page on real-time is pretty good for definitions. – robert bristow-johnson Sep 24 '18 at 7:34
• Can I please ask if this was resolved? – A_A Sep 28 '18 at 11:34
• I was asking this question with respect to inference time in a deep learning based system, which uses image input. – Ajinkya Ghadge Oct 4 '18 at 4:51

There are as many ad hoc "definitions" of "real-time" are there are people or places you hear or read about what it supposedly means. Some people claim there is a "hardline" such as 10uS response time, but I believe there is no academic paper that makes that mistake.

Many people implicitly have an informal mental model that considers a system as being "real-time" • if, or to the extent that, • its actions are manifest to action observers with latencies that can be related (usually in an application-specific way) to the actions’ perceived currencies— • i.e., in a time frame that those actions’ latencies and predictability of latencies have acceptably satisfactory value to their observers (e.g., system users). Think of high frequency trading.

The magnitudes of the time frames are application- and situation-specific—e.g., microseconds to megaseconds—not related to if, or to the extent, that a system is a real-time one.

So a system "operates in real-time" if its actions have low enough latency to satisfy your needs. If 15 FPS is a satisfactory enough rate, then the system is real-time as far as your needs are concerned; otherwise it is not--that's all that matters to you.

• Hi mr. Jensen! The first time I heard real-time (many years ago), I was an intern and there was a guy doing RTOS application presentation. And he asked me "do you know what an RTOS is ?" I said, "hmm well think about a (inverted) pendulum, you want to control it and send signals to it, you must get its position as soon as possible and send back the control signal asap, otherwise it's too late, something like that isn't it? He said, "well it's not"... – Fat32 Oct 1 '18 at 22:34
• @Fat32 You answered with a description of a real-time control application. I assume the other person was asking you for a description of an RTOS. The software to perform the inverted pendulum control operations must satisfy the application-specific timeliness and predictability of timeliness requirements. An RTOS is not required for that but may be helpful. There are numerous ad hoc attempts at defining an "RTOS" on the web and elsewhere. Few are precise and complete. – E. Douglas Jensen Oct 3 '18 at 10:03

"Real Time" is defined with respect to the problem. Typically, "Real Time" is any system that carries out any computation required to provide an output in time less than the sampling period $$Ts$$.

To define "Real Time" for your application, you need a quantification of "time" or "flow" or "rate" so that you can compare that to what your system achieves or should be achieving. You need to determine how many decisions per unit of time should the classifier be making to decide if it would be adequate for real time operation.

Hope this helps.