Spaces:
Sleeping
Sleeping
File size: 2,729 Bytes
0cef02f 2d6fc22 0cef02f 9b1ccc9 80a5ad2 2d6fc22 0cef02f 9b1ccc9 2d6fc22 9cbc798 2d6fc22 9cbc798 2d6fc22 9332803 f1e3f4b 9332803 f1e3f4b 9332803 f1e3f4b 9332803 9b1ccc9 2ec7210 80a5ad2 9332803 |
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 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 |
import gradio as gr
import cv2
import torch
import numpy as np
from diffusers import StableDiffusionPipeline,AutoPipelineForImage2Image,AutoencoderTiny
from transformers import AutoProcessor, AutoModel, AutoTokenizer
from PIL import Image
#
#
# def predict(prompt, frame):
# generator = torch.manual_seed(params.seed)
# steps = params.steps
# strength = params.strength
# if int(steps * strength) < 1:
# steps = math.ceil(1 / max(0.10, strength))
#
# prompt = params.prompt
# prompt_embeds = None
#
# results = self.pipe(
# image=frame,
# prompt_embeds=prompt_embeds,
# prompt=prompt,
# negative_prompt=params.negative_prompt,
# generator=generator,
# strength=strength,
# num_inference_steps=steps,
# guidance_scale=1.1,
# width=params.width,
# height=params.height,
# output_type="pil",
# )
#
# nsfw_content_detected = (
# results.nsfw_content_detected[0]
# if "nsfw_content_detected" in results
# else False
# )
# if nsfw_content_detected:
# return None
# result_image = results.images[0]
#
# return result_image
#
# def process_frame(frame, prompt="A futuristic landscape"):
# """Process a single frame using the real-time latent consistency model."""
#
# # Convert frame to PIL image
# image = Image.fromarray(cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)).resize((512, 512))
#
# # Apply Real-Time Latent Consistency Model
# result = realtime_pipe(prompt=prompt, image=image, strength=0.5, guidance_scale=7.5).images[0]
# return np.array(result)
def video_stream(prompt):
"""Captures video feed from webcam and sends to the AI model."""
cap = cv2.VideoCapture(0)
while cap.isOpened():
ret, frame = cap.read()
if not ret:
break
frame = process_frame(frame, prompt)
yield frame # Return processed frame
cap.release()
# Create Gradio App
with gr.Blocks() as demo:
gr.Markdown("## 🎨 Real-Time AI-Enhanced Webcam using Latent Consistency Model (LCM)")
with gr.Row():
webcam_feed = gr.Camera(streaming=True, label="Live Webcam")
processed_image = gr.Image(label="AI-Enhanced Webcam Feed")
with gr.Row():
canvas_output = gr.Image(interactive=True, label="Canvas - Processed Image Output")
prompt_input = gr.Textbox(label="Real-Time LCM Prompt", value="A futuristic landscape")
start_button = gr.Button("Start Real-Time AI Enhancement")
#start_button.click(fn=video_stream, inputs=[prompt_input], outputs=[processed_image, canvas_output])
demo.launch(share=True)
|