APopTrainer / app.py
sschmied's picture
Upload app.py
0736f12 verified
Raw History Blame Contribute Delete
3.91 kB
# app.py
import gradio as gr
import cv2
import mediapipe as mp
import numpy as np
import plotly.graph_objects as go
import math
import logging
# Initialize MediaPipe Pose model
mp_pose = mp.solutions.pose
pose = mp_pose.Pose(
static_image_mode=True,
model_complexity=2,
min_detection_confidence=0.5
)
# Browser TTS Integration
tts_js = """
<script>
function speak(text) {
if (window.speechSynthesis) {
const utterance = new SpeechSynthesisUtterance(text);
utterance.volume = 1;
utterance.rate = 1;
window.speechSynthesis.speak(utterance);
}
}
// Activate TTS on any button click
document.addEventListener("DOMContentLoaded", () => {
document.querySelectorAll('button').forEach(button => {
button.addEventListener('click', () => {
if (window.speechSynthesis) {
// Pre-activate audio context
window.speechSynthesis.cancel();
}
});
});
});
</script>
"""
def calculate_angle(a, b, c):
ba = np.array(a) - np.array(b)
bc = np.array(c) - np.array(b)
cosine_angle = np.dot(ba, bc) / (np.linalg.norm(ba) * np.linalg.norm(bc))
return np.degrees(np.arccos(np.clip(cosine_angle, -1.0, 1.0)))
def process_image(img):
try:
results = pose.process(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
output_image = np.zeros_like(img)
if results.pose_landmarks:
# Draw pose landmarks
mp.solutions.drawing_utils.draw_landmarks(
output_image,
results.pose_landmarks,
mp_pose.POSE_CONNECTIONS,
landmark_drawing_spec=mp.solutions.drawing_styles.get_default_pose_landmarks_style()
)
# Get key points
landmarks = results.pose_landmarks.landmark
nose = landmarks[mp_pose.PoseLandmark.NOSE.value]
left_hip = landmarks[mp_pose.PoseLandmark.LEFT_HIP.value]
right_hip = landmarks[mp_pose.PoseLandmark.RIGHT_HIP.value]
# Calculate angles
vertical_angle = math.degrees(math.atan2(
left_hip.x - right_hip.x,
left_hip.y - right_hip.y
))
# Create feedback message
feedback = []
if vertical_angle > 15:
feedback.append("Keep your hips lower")
if nose.y > 0.8:
feedback.append("Look forward")
feedback_html = "<br>".join(feedback) if feedback else "Good form!"
else:
feedback_html = "No pose detected"
output_image = img
return cv2.cvtColor(output_image, cv2.COLOR_BGR2RGB), feedback_html
except Exception as e:
logging.error(f"Processing error: {e}")
return img, "Error processing image"
def build_interface():
with gr.Blocks(title="Surf Trainer", css=".gradio-container {background: #f0f0f0}") as app:
gr.HTML(tts_js)
gr.Markdown("# 🏄 Surf Pop-Up Trainer")
with gr.Row():
with gr.Column():
image_input = gr.Image(label="Input Image", sources=["upload"])
submit_btn = gr.Button("Analyze", variant="primary")
with gr.Column():
image_output = gr.Image(label="Pose Analysis", interactive=False)
feedback_output = gr.HTML(label="Feedback")
submit_btn.click(
fn=process_image,
inputs=image_input,
outputs=[image_output, feedback_output]
)
gr.Markdown("---\n<div style='text-align: center'>Powered by MediaPipe & Gradio</div>")
return app
if __name__ == "__main__":
build_interface().launch()