The LiveVBR open-source code is developed by Christian Doppler Laboratory ATHENA, University of Klagenfurt, Austria. The primary objective of LiveVBR is a two-pass cVBR encoding scheme with a content-adaptive, JND-aware, online bitrate ladder prediction optimized for adaptive live streaming applications. The minimum and maximum encoding bitrates, the maximum quality level, and the target average JND function are considered as inputs to the scheme. Moreover, the encoder/codec used, is input to the scheme to ensure that the bitrate ladder is generated for the corresponding encoder. Based on the video complexity features (extracted by VCA) and the input parameters, bitrate-resolution-CRF triples are predicted. The adjacent points of the bitrate ladder are envisioned to have a perceptual quality difference of one JND. Although reducing the overall storage needed to store the representations, LiveVBR is expected to improve the overall compression efficiency of the bitrate ladder encoding. LiveVBR is available as an open source python code, published under the GPLv3 license.
