I test AR and video SDKs on real phones and write down what happens. Face tracking, filters, beauty AR, background removal, mobile video editors. Mostly Android and iOS, some React Native and Flutter.
Every post carries the method: device, OS version, SDK build, what I measured and how. That way you can repeat it and tell me I got it wrong.
I publish the runs that went badly too. A benchmark that only ever finds good news is not a benchmark.
TL;DR For augmented reality live streaming, the AR SDK shares the device with the video encoder, so treat the frame rate a vendor publishes for an idle phone as a ceiling. DeepAR publishes the most u
TL;DR A background removal API can score beautifully on still frames and look broken in motion, because per-frame accuracy and frame-to-frame stability are separate properties and only one of them ge
The sales call and the reference manual describe two different products. On the call you get lipstick sitting perfectly on a face somebody lit properly. In the manual you find out whether eyeliner is
