Video coding optimization with dynamic optimizer
keywords MULTIMEDIA, OPTIMISATION, VIDEO COMPRESSION, VIDEO QUALITY
Reference persons ENRICO MASALA
Research Groups INTERNET MEDIA GROUP - IMG
Thesis type RESEARCH / EXPERIMENTAL
Description In the context of the research activities of the Internet Media Group, we are looking for a master thesis student with strong interests in investigating an innovative video coding optimization framework similar to the one described on the Netflix techblog https://netflixtechblog.com/dynamic-optimizer-a-perceptual-video-encoding-optimization-framework-e19f1e3a277f . The student will implement the technique and evaluate its performance in terms of computational complexity and visual quality using state of the art measures such as Netflix VMAF comparing it to more traditional coding schemes. The activity, if needed, can leverage servers available to the resaerch group. The activity does not require to be physically present at the Politecnico.
See also https://media.polito.it/wordpress/theses/specific-proposals/thesis-video-coding-optimization-with-dynamic-optimizer/index.html
Required skills Reasonable knowledge of Unix/Linux (basic shell scripting); Reasonable knowledge of digital multimedia signal coding tools, i.e., being able to encode/decode a video and evaluate the resulting quality. A reasonably good mark in the "Internet Video Streaming" course (or the older course "Elaborazione e trasmissione di informazioni multimediali" course) is strongly appreciated but not strictly required if the student is strongly motivated.
Deadline 29/03/2022 PROPONI LA TUA CANDIDATURA