-
Notifications
You must be signed in to change notification settings - Fork 0
/
app.py
25 lines (21 loc) · 1.09 KB
/
app.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
import streamlit as st
import cv2
import matplotlib.pyplot as plt
import pixellib
from pixellib.semantic import semantic_segmentation
from prediction import *
st.title("Instance Image Segmentation")
st.sidebar.title("Instance Image Segmentation")
st.sidebar.markdown("Mask R-CNN model trained on coco dataset for Instance Segmentation using PixelLib. Here, Different instances of the same object are segmented with the different color maps.")
pic = plt.imread("coco_dataset.png")
st.sidebar.image(pic, caption = "Model classes", use_column_width = True)
st.set_option('deprecation.showfileUploaderEncoding', False)
img_file = st.file_uploader("Upload the input image : ", type = ['jpg', 'jpeg', 'png'])
if img_file is not None:
img = plt.imread(img_file, 0)
st.image(img, caption = "Input Image", use_column_width = True)
cv2.imwrite("input_images/in.jpg", img)
out, out_box = prediction("input_images/in.jpg")
col1, col2 = st.beta_columns(2)
col1.image(out, caption = "Segmented Image", use_column_width = True)
col2.image(out_box, caption = "Bounding Box", use_column_width = True)