|
|
import gradio as gr |
|
|
import matplotlib |
|
|
from matplotlib import gridspec |
|
|
import matplotlib.pyplot as plt |
|
|
import numpy as np |
|
|
from PIL import Image |
|
|
import torch |
|
|
from transformers import AutoImageProcessor, AutoModelForSemanticSegmentation |
|
|
import time |
|
|
|
|
|
|
|
|
MODEL_ID = "nvidia/segformer-b2-finetuned-cityscapes-1024-1024" |
|
|
processor = AutoImageProcessor.from_pretrained(MODEL_ID) |
|
|
model = AutoModelForSemanticSegmentation.from_pretrained(MODEL_ID) |
|
|
|
|
|
def ade_palette(): |
|
|
"""ADE20K palette that maps each class to RGB values.""" |
|
|
return [ |
|
|
[128, 64, 128], [244, 35, 232], [70, 70, 70], [102, 102, 156], [190, 153, 153], |
|
|
[153, 153, 153], [250, 170, 30], [220, 220, 0], [107, 142, 35], [152, 251, 152], |
|
|
[70, 130, 180], [220, 20, 60], [255, 0, 0], [0, 0, 142], [0, 0, 70], |
|
|
[0, 60, 100], [0, 80, 100], [0, 0, 230], [119, 11, 32] |
|
|
] |
|
|
|
|
|
|
|
|
labels_list = [] |
|
|
with open("labels.txt", "r", encoding="utf-8") as fp: |
|
|
for line in fp: |
|
|
labels_list.append(line.rstrip("\n")) |
|
|
|
|
|
colormap = np.asarray(ade_palette(), dtype=np.uint8) |
|
|
|
|
|
def label_to_color_image(label): |
|
|
if label.ndim != 2: |
|
|
raise ValueError("Expect 2-D input label") |
|
|
if np.max(label) >= len(colormap): |
|
|
raise ValueError("label value too large.") |
|
|
return colormap[label] |
|
|
|
|
|
|
|
|
def draw_plot(pred_img, seg_np): |
|
|
|
|
|
fig = plt.figure(figsize=(25, 20)) |
|
|
|
|
|
|
|
|
grid_spec = gridspec.GridSpec(1, 2, width_ratios=[8, 1]) |
|
|
|
|
|
plt.subplot(grid_spec[0]) |
|
|
plt.imshow(pred_img) |
|
|
plt.axis('off') |
|
|
|
|
|
LABEL_NAMES = np.asarray(labels_list) |
|
|
FULL_LABEL_MAP = np.arange(len(LABEL_NAMES)).reshape(len(LABEL_NAMES), 1) |
|
|
FULL_COLOR_MAP = label_to_color_image(FULL_LABEL_MAP) |
|
|
|
|
|
unique_labels = np.unique(seg_np.astype("uint8")) |
|
|
|
|
|
valid_labels = [label for label in unique_labels if label < len(LABEL_NAMES)] |
|
|
|
|
|
ax = plt.subplot(grid_spec[1]) |
|
|
plt.imshow(FULL_COLOR_MAP[valid_labels].astype(np.uint8), interpolation="nearest") |
|
|
ax.yaxis.tick_right() |
|
|
plt.yticks(range(len(valid_labels)), LABEL_NAMES[valid_labels]) |
|
|
plt.xticks([], []) |
|
|
ax.tick_params(width=0.0, labelsize=25) |
|
|
return fig |
|
|
|
|
|
|
|
|
def run_inference(input_img, alpha=0.5): |
|
|
start_time = time.time() |
|
|
img = Image.fromarray(input_img.astype(np.uint8)) if isinstance(input_img, np.ndarray) else input_img |
|
|
if img.mode != "RGB": |
|
|
img = img.convert("RGB") |
|
|
|
|
|
inputs = processor(images=img, return_tensors="pt") |
|
|
|
|
|
with torch.no_grad(): |
|
|
outputs = model(**inputs) |
|
|
logits = outputs.logits |
|
|
|
|
|
upsampled = torch.nn.functional.interpolate( |
|
|
logits, size=img.size[::-1], mode="bilinear", align_corners=False |
|
|
) |
|
|
seg = upsampled.argmax(dim=1)[0].cpu().numpy().astype(np.uint8) |
|
|
|
|
|
color_seg = colormap[seg] |
|
|
|
|
|
|
|
|
image_weight = 1.0 - alpha |
|
|
overlay_weight = alpha |
|
|
pred_img = (np.array(img) * image_weight + color_seg * overlay_weight).astype(np.uint8) |
|
|
|
|
|
fig = draw_plot(pred_img, seg) |
|
|
print(f"Inference time: {time.time() - start_time:.2f}s") |
|
|
return fig |
|
|
|
|
|
|
|
|
custom_theme = gr.themes.Soft( |
|
|
primary_hue="emerald", |
|
|
secondary_hue="teal", |
|
|
neutral_hue="slate" |
|
|
).set( |
|
|
body_background_fill="#0f172a", |
|
|
body_text_color="#e2f1e8", |
|
|
button_primary_background_fill="#10b981", |
|
|
button_primary_text_color="#ffffff", |
|
|
block_background_fill="#1a2e25", |
|
|
) |
|
|
|
|
|
demo = gr.Interface( |
|
|
fn=run_inference, |
|
|
|
|
|
|
|
|
inputs=[ |
|
|
gr.Image(type="numpy", label="๐ธ Input Image"), |
|
|
gr.Slider(0.0, 1.0, value=0.5, step=0.05, label="Overlay Transparency (ํฌ๋ช
๋)") |
|
|
], |
|
|
|
|
|
outputs=gr.Plot(label="Overlay + Legend"), |
|
|
|
|
|
|
|
|
examples=[ |
|
|
["city1.png", 0.5], |
|
|
["city2.png", 0.5], |
|
|
["city3.jpg", 0.5], |
|
|
["city4.jpeg", 0.5], |
|
|
["city5.jpg", 0.5] |
|
|
], |
|
|
|
|
|
flagging_mode="never", |
|
|
cache_examples=False, |
|
|
title="๐๏ธ City Segment", |
|
|
description=( |
|
|
"segformer-b2๋ชจ๋ธ์ ์ด์ฉ ๋์ ์ด๋ฏธ์ง ๋ถํ ์๊ฐ.<br>" |
|
|
"์ด๋ฏธ์ง๋ฅผ ์
๋ก๋ํ๋ฉด ๋๋ก, ๊ฑด๋ฌผ, ์ฐจ๋, ์ฌ๋ ๋ฑ ๊ฐ์ฒด๋ณ๋ก ์์์ผ๋ก ๊ตฌ๋ถํด์ค๋๋ค." |
|
|
), |
|
|
theme=custom_theme |
|
|
) |
|
|
|
|
|
if __name__ == "__main__": |
|
|
demo.launch() |