Releases: duckysmacky/signsense
v1.2.0
Recognition update
This version improved the algorithm behind recognising signs and translating them into letters by adding a "Recognition Threshold". This means that once a sign was found, it has to pass a specific threshold, which is based on scores gotten after running found landmarks via the Pytorch model. It can be configured in settings and by default set to 80% (which translated to >= 20 recognition score). Setting it above 80% is not reccomended, as the app might not translate the sign at all.
Now once the sign was recognised and it has passed the Recognition Threshold, it is immidiatly added to translated letters. Duplicate letters coming after each other (e.g.: "aabaa") in the translated word are not allowed any more (due to new recognition system).
There have also been many improvements to the camera inself, with slight optimisations (+-10 fps on average -> +-18 fps on average) and fixes to many bugs which appeared on the Release build of the app (not present on Debug build).
Changes
- Fixed bug where on Release build OpenCV wouldn't open
- General optimisations
- FPS counter setting
- Sign recognition and translation algorithm overhaul
- Now there is a Recognition Threshold which has to be passed in order to register a sign as translated (configurable in settings, 80% by default)
- Sign is immidiatly registered and added to translated letters upon passing the threshold (no delay)
- Removed duplicate letter translation
- Color coded hand sceleton based on recognition status (not recognised - red, recognised - green and new recognition - blue)
Installation
- Download the apk file from here or in assets below
- Open and install the apk file
- Done
v1.1.0
Optimisation update
Due to previously big apk size, I decided to run some optimisations on the app. After many painful hours configuring Proguard, Multidex and Native Libraries I was able to reduce its size by 70%. I am planning to reduce the size even more in the future in order to provide easier app installation for users.
Because of the reduced apk size, I can now upload the app directly into Github's releases section, but Mediafire is still an alternative source for downloading the app.
Changes
- Reduced apk size (~160mb)
- Faster app startup speeds
- Overall optimisations
Installation
- Download the apk file from here or in assets below
- Open and install the apk file
- Done
v1.0.0 - Release
Full release
After being in development for over 2 months, the app is finally ready for release.
Installation guide:
The app was tested on Android versions 8, 9, 10, 11 and 12, min supported version is Android 8.
To install the app and try it out yourself:
- Download the APK file below (if your browser is blocking you from downloading, enable download from unknown sources feature in settings)
- Click on the file, press install
- Wait for the installation to finish and you are done!
Bugs and feature requests
The app might still crash and have several bugs as this release was a bit rushed because of our project presentation. Make sure to submit all bugs to issues, I will try to fix them as soon as possible. Any other feature requests can also be submited there, as well as directly to me (see profile README).
The app itself is pretty big (around 450mb) because of different libraries (such as OpenCV, Mediapipe and PyTorch) used in the project