Our supplementary materials contain two files.
1. '12611_supp.mp4': the video that demonstrates our motivation, the proposed generic avatar editing approach, and also shows results of expression-aware avatar editing.
2. '12611_supp.pdf': the pdf file describes more details of our method, discussions, and more experiments.
	
Our video is encoded with the FFmpeg encoder x264, and you can find the codecs through the following https://trac.ffmpeg.org/wiki/Encode/H.264 . 