Proof of concept for generating a high-resolution video with raw binary image files piped to ffmpeg

Using specific flags to get ffmpeg into a mode where it reads binary bytes of frames in RGB32 and assembles them into a video in real-time

Idea: ffmpeg should be started as a child_process by the program together with a pipe where it gets fed raw data from the program. Ffmpeg then outputs assembled video.

Quick Start

git clone [email protected]:toxypiks/neural_network_video_example.git

cd neural_network_video_example

mkdir build

cd build

cmake ..

make

./simple_example