Commit
·
ab123cf
1
Parent(s):
d280220
Update app.py
Browse files
app.py
CHANGED
@@ -332,7 +332,21 @@ def translate_sign_language(gesture):
|
|
332 |
|
333 |
with gr.Blocks(css=custom_css) as demo:
|
334 |
gr.Markdown("# Indian Sign Language Translation App")
|
335 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
336 |
# Gesture recognition Tab
|
337 |
with gr.Tab("Gesture recognition"):
|
338 |
|
|
|
332 |
|
333 |
with gr.Blocks(css=custom_css) as demo:
|
334 |
gr.Markdown("# Indian Sign Language Translation App")
|
335 |
+
|
336 |
+
# About the App
|
337 |
+
with gr.Tab("About the App"):
|
338 |
+
gr.Markdown("""
|
339 |
+
# Product Description/Objective
|
340 |
+
Our innovative AI solution bridges the communication gap between the hearing-impaired community and the wider society. Designed to recognize and translate 75 hand signs and gestures from Indian Sign Language into English words,
|
341 |
+
this technology ensures seamless interaction. Additionally, the product includes animated videos that visually demonstrate the hand signs and gestures corresponding to each English word, fostering inclusivity and accessibility in communication.
|
342 |
+
# How does it work
|
343 |
+
Our solution uses advanced AI technology to process videos and identify sign language gestures. The process begins by extracting pose coordinates, which include the positions of the hands, face, and body edges, from each frame of the video.
|
344 |
+
These coordinates act as a blueprint of the movements and gestures performed by the person in the video. By analyzing these hand movements in detail, the AI model identifies the gestures being made and matches them to the most likely English word associated with that specific sign.
|
345 |
+
For instance, the image provided illustrates a person performing the gesture for the word "Student," demonstrating the system's ability to interpret and translate sign language gestures into meaningful English words.
|
346 |
+
In addition to recognizing gestures, our solution also visualizes them through animated motion videos. During the AI model's training phase, face and hand coordinates are collected from the videos to create dynamic animations that represent all the sign language gestures the AI has learned to recognize.
|
347 |
+
These animations serve as a visual guide, making it easier for users to understand and learn sign language gestures. For example, the motion video below demonstrates how the word "Student" is represented in sign language, showcasing both the accuracy and clarity of the system's animated outputs.
|
348 |
+
""")
|
349 |
+
|
350 |
# Gesture recognition Tab
|
351 |
with gr.Tab("Gesture recognition"):
|
352 |
|