cneidle
commented
1 year ago This is not quite the right distinction.
What you describe as “non-continuous” applies to our ASLLVD data.
We also have “non-continuous” data from RIT and DSP, which do not have any composite videos, but there is also no utterance to display.
On Sep 22, 2022, at 10:11 AM, Greg Dimitriadis @.***> wrote:
Background information:
The data used for SignStream - i.e. the collections, photos, videos, etc - classify two types of signs: continuous and non-continuous. What these refer to it not important, but the difference as far as SignStream is concerned is this: Non-continuous signs in the global sign bank file have 2 videos, both are of TYPE='SIGN' (the second is a "composite" video of many videos stitched together) Continuous signs in the global sign bank file have 2 videos, one is TYPE='SIGN' and one is TYPE='UTTERANCE' In SignStream, on the sign bank variant result panel, there are three buttons, two of which play a video. This issue is made because for signs that are continuous, the text for the button would say "Play Composite Video' but composites do not apply. So we should instead display "Play Utterance Video" For now, in the current code, a quick fix was done to just say "Play Composite/Utterance Video" to work for either scenario. But now we want to detect which type of video it is and put the right text on the button
Todos:
(first, note: there is already an arraylist meant for utterance videos in the SignBankUtterance class, just unused mostly) change the sign bank parser code to detect the TYPE of a video, and add the video to the proper list - either sign or utterance make sure this never fails for existing signs bubble up the code everywhere necessary up to the sign bank variant panel result dislay
— Reply to this email directly, view it on GitHub https://github.com/DCS-LCSR/SignStream3/issues/554, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADH7NQSZHC7S6JLFOESXEJDV7RSHVANCNFSM6AAAAAAQTDCQFY. You are receiving this because you are subscribed to this thread.
Background information:
The data used for SignStream - i.e. the collections, photos, videos, etc - classify two types of signs: continuous and non-continuous. What these refer to it not important, but the difference as far as SignStream is concerned is this: Non-continuous signs in the global sign bank file have 2 videos, both are of TYPE='SIGN' (the second is a "composite" video of many videos stitched together) Continuous signs in the global sign bank file have 2 videos, one is TYPE='SIGN' and one is TYPE='UTTERANCE' In SignStream, on the sign bank variant result panel, there are three buttons, two of which play a video. This issue is made because for signs that are continuous, the text for the button would say "Play Composite Video' but composites do not apply. So we should instead display "Play Utterance Video" For now, in the current code, a quick fix was done to just say "Play Composite/Utterance Video" to work for either scenario. But now we want to detect which type of video it is and put the right text on the button
Todos:
(first, note: there is already an arraylist meant for utterance videos in the SignBankUtterance class, just unused mostly) change the sign bank parser code to detect the TYPE of a video, and add the video to the proper list - either sign or utterance make sure this never fails for existing signs bubble up the code everywhere necessary up to the sign bank variant panel result dislay