Update app.py
Browse files
app.py
CHANGED
|
@@ -3,17 +3,21 @@ import torch
|
|
| 3 |
from transformers import AutoImageProcessor, VideoMAEForVideoClassification
|
| 4 |
import cv2
|
| 5 |
import numpy as np
|
|
|
|
| 6 |
|
| 7 |
# Load model and processor
|
| 8 |
model = VideoMAEForVideoClassification.from_pretrained("model")
|
| 9 |
processor = AutoImageProcessor.from_pretrained("model")
|
| 10 |
model.eval()
|
| 11 |
|
| 12 |
-
# Function to extract and preprocess 16
|
| 13 |
def preprocess_video(video_path, frame_count=16):
|
| 14 |
cap = cv2.VideoCapture(video_path)
|
| 15 |
frames = []
|
| 16 |
total = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
|
|
|
|
|
|
|
|
|
|
| 17 |
for i in np.linspace(0, total - 1, frame_count, dtype=int):
|
| 18 |
cap.set(cv2.CAP_PROP_POS_FRAMES, i)
|
| 19 |
ret, frame = cap.read()
|
|
@@ -26,20 +30,22 @@ def preprocess_video(video_path, frame_count=16):
|
|
| 26 |
|
| 27 |
# Inference function
|
| 28 |
def predict(video_path):
|
|
|
|
|
|
|
| 29 |
frames = preprocess_video(video_path)
|
| 30 |
-
if
|
| 31 |
-
return "❌
|
| 32 |
inputs = processor(frames, return_tensors="pt")
|
| 33 |
with torch.no_grad():
|
| 34 |
logits = model(**inputs).logits
|
| 35 |
predicted = torch.argmax(logits, dim=1).item()
|
| 36 |
-
return "⚠️ Abnormal" if predicted == 1 else "✅ Normal"
|
| 37 |
|
| 38 |
-
#
|
| 39 |
gr.Interface(
|
| 40 |
fn=predict,
|
| 41 |
-
inputs=gr.Video(label="Upload
|
| 42 |
-
outputs="
|
| 43 |
-
title="Abnormal Behavior Detection"
|
|
|
|
| 44 |
).launch()
|
| 45 |
-
|
|
|
|
| 3 |
from transformers import AutoImageProcessor, VideoMAEForVideoClassification
|
| 4 |
import cv2
|
| 5 |
import numpy as np
|
| 6 |
+
import os
|
| 7 |
|
| 8 |
# Load model and processor
|
| 9 |
model = VideoMAEForVideoClassification.from_pretrained("model")
|
| 10 |
processor = AutoImageProcessor.from_pretrained("model")
|
| 11 |
model.eval()
|
| 12 |
|
| 13 |
+
# Function to extract and preprocess 16 frames
|
| 14 |
def preprocess_video(video_path, frame_count=16):
|
| 15 |
cap = cv2.VideoCapture(video_path)
|
| 16 |
frames = []
|
| 17 |
total = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
|
| 18 |
+
if total < frame_count:
|
| 19 |
+
cap.release()
|
| 20 |
+
return None
|
| 21 |
for i in np.linspace(0, total - 1, frame_count, dtype=int):
|
| 22 |
cap.set(cv2.CAP_PROP_POS_FRAMES, i)
|
| 23 |
ret, frame = cap.read()
|
|
|
|
| 30 |
|
| 31 |
# Inference function
|
| 32 |
def predict(video_path):
|
| 33 |
+
if video_path is None or not os.path.exists(video_path):
|
| 34 |
+
return "❌ No video uploaded or camera input failed."
|
| 35 |
frames = preprocess_video(video_path)
|
| 36 |
+
if frames is None:
|
| 37 |
+
return "❌ Couldn't read video or not enough frames."
|
| 38 |
inputs = processor(frames, return_tensors="pt")
|
| 39 |
with torch.no_grad():
|
| 40 |
logits = model(**inputs).logits
|
| 41 |
predicted = torch.argmax(logits, dim=1).item()
|
| 42 |
+
return "⚠️ Abnormal Behavior Detected" if predicted == 1 else "✅ Normal Behavior"
|
| 43 |
|
| 44 |
+
# Interface
|
| 45 |
gr.Interface(
|
| 46 |
fn=predict,
|
| 47 |
+
inputs=gr.Video(label="Upload .mp4 video or record with webcam", format="mp4"),
|
| 48 |
+
outputs=gr.Text(label="Prediction"),
|
| 49 |
+
title="Abnormal Behavior Detection",
|
| 50 |
+
description="Upload a video in .mp4 format or use your webcam to analyze behavior"
|
| 51 |
).launch()
|
|
|