3 ms·
By generating the "code" are you merely inferring / representing UI layout as some kind of a tree. E.g. each button/panel/textbox gets detected, and an "RNN/LST
by aub3bhat 9y ago
By generating the "code" are you merely inferring / representing UI layout as some kind of a tree. E.g. each button/panel/textbox gets detected, and an "RNN/LSTM" generate a tree structure using attention.
- avg_dev 9y ago"merely" is not the word I would use.
- tanilama 9y agoIf you somewhat familiar with DL literature, you will see this paper, while having a very interesting angle, the underlying architecture is a standard, enc-dec network, with encoder being CNN and decoding being LSTM. Such application, has been studied before: https://arxiv.org/pdf/1609.04938.pdf https://arxiv.org/pdf/1609.04938.pdf The above paper shows nice result that turns image to latex expression, and image to html.