Create a RAII object which measures CPU and optionally GPU time for the loop which calls decode() method

Decrement the ref.counter, and return true if it reached zero as the result

naive Discrete Fourier Transform input is real-valued output is complex-valued

naive Discrete Fourier Transform input is real-valued output is complex-valued

a / b, rounded up to the next integer

[ x, y ] => [ x, y, x, y ]

Flip order of bytes from RGB to BGR, or vice versa

<summary>Full the default parameters of the model, for the specified sampling strategy</summary>

Find audio decoder MFT, query MF_MT_SUBTYPE attribute of the current input media type of that MFT

A vector of 2 uint64_t values, both numbers are 100 nanosecond ticks: 0. The time it took to load the model, measured on CPU 1. The time it took to up