vu0018's picture
Update app.py
af17b79 verified
raw
history blame
3.43 kB
import gradio as gr
import cv2
import mediapipe as mp
import torch
import numpy as np
import tempfile
# Initialize MediaPipe Pose
mp_pose = mp.solutions.pose
# Dummy ST-GCN++ model (replace with actual model)
class SimpleSTGCNPlusPlus(torch.nn.Module):
def __init__(self, input_size=99, num_classes=5): # 33 keypoints x 3 coords
super().__init__()
self.fc = torch.nn.Sequential(
torch.nn.Linear(input_size, 64),
torch.nn.ReLU(),
torch.nn.Linear(64, num_classes)
)
def forward(self, x):
return self.fc(x)
# Instantiate the model
model = SimpleSTGCNPlusPlus()
labels = ["Ballet Dancing", "Cycling", "Running", "Jumping", "Walking"]
def detect_pose_and_activity(video_file):
try:
# Save uploaded video to a temporary file
temp_video = tempfile.NamedTemporaryFile(delete=False, suffix=".mp4")
temp_video.write(open(video_file, "rb").read())
temp_video.close()
cap = cv2.VideoCapture(temp_video.name)
fps = cap.get(cv2.CAP_PROP_FPS)
total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
duration = total_frames / fps
max_frames = int(min(duration, 10) * fps)
output_frames = []
keypoints_sequence = []
with mp_pose.Pose(static_image_mode=False, min_detection_confidence=0.5) as pose:
frame_count = 0
while frame_count < max_frames:
ret, frame = cap.read()
if not ret:
break
image_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
results = pose.process(image_rgb)
if results.pose_landmarks:
keypoints = []
for lm in results.pose_landmarks.landmark:
keypoints.extend([lm.x, lm.y, lm.z])
keypoints_sequence.append(keypoints)
mp.solutions.drawing_utils.draw_landmarks(frame, results.pose_landmarks, mp_pose.POSE_CONNECTIONS)
else:
keypoints_sequence.append([0] * 99)
output_frames.append(frame)
frame_count += 1
cap.release()
if not keypoints_sequence:
return None, "No pose detected."
keypoints_tensor = torch.tensor(keypoints_sequence, dtype=torch.float32).mean(dim=0, keepdim=True)
with torch.no_grad():
preds = model(keypoints_tensor)
action_idx = torch.argmax(preds, dim=1).item()
action_label = labels[action_idx]
output_file = tempfile.NamedTemporaryFile(delete=False, suffix=".mp4").name
height, width, _ = output_frames[0].shape
out = cv2.VideoWriter(output_file, cv2.VideoWriter_fourcc(*'mp4v'), fps, (width, height))
for f in output_frames:
out.write(f)
out.release()
return output_file, f"Predicted Action: {action_label}"
except Exception as e:
return None, f"Error during processing: {str(e)}"
iface = gr.Interface(
fn=detect_pose_and_activity,
inputs=gr.Video(label="Upload a Video (max 10s)"),
outputs=[gr.Video(label="Pose Detection Output"), gr.Textbox(label="Detected Action")],
title="Human Pose & Activity Recognition",
description="Upload a short video, and this app will detect human poses and predict the activity (e.g., ballet, cycling)."
)
iface.launch()