Goal of this project was to create multiple models for detecting sign language gestures and translate them into text. Additional a demo application was created, to provide a graphical user interface to use the models.
For more details see: Projektdokumentation.pdf
Repo tested on
- Ubuntu 22.04
- Cuda 11.7
- Conda 23.3.1
The environments can be used for multiple stages with minimal changes. After those adjustements it can still be used for the original intended stage.
| Step | Environment Setup |
|---|---|
| 01_data_recording | recording/readme |
| 02_data_preprocessing | recording/readme |
| 03_training | training/readme |
| 05_demo | demo/readme |
| 07_finetuning | finetuning/readme |