" $ , 5 0 / ) TU " TU $ 0 , ) OE 5 " OE / Algorithm Instruction &YBNQMFPGBMHPSJUIN We use an auto-encoding CNN and a seq2seq LSTM for efficient compression.
time and GPU 5PUBMMZOPCPVOEBSJFTGPSBOZPOFPWFS WJEFPT Future Certain divide around us over video accessibility and quality $PNJOH Better infrastructure to fill in the gap ɾ ɾ ɾ
Reasoning HD (High Definition) > 2.5 Mbps Minimum goodput required to sustain an average YouTube HD video playback at 720p resolution SD (Standard Definition) 0.7 to 2.5 Mbps Minimum goodput required to sustain an average YouTube SD video playback at 360p resolution LD (Lower Definition) < 0.7 Mbps Goodput too low to sustain YouTube SD video playback at 360p resolution
compressed video We help invalidate digital divide over video usage, by making an application with AI technology that reduces the video size up to 95%. Our prioritized market is people in 40 developing countries. ɾHow well fit it is to the theme(co-living of humans and AI) and creativity ɾLevel of completion Our Core Messages to Support the Value Proposition ɾWe shoudn’t think AI v humans; we prefer a world where AI is woven in our lives that we don’t even have to dinsinguish AI ɾWe have made it possible to compress 1min video to 5% for 30 seconds the within 48 hours, and it is highly possible to Executive Summary