Spaces:
Running
on
Zero
Running
on
Zero
File size: 5,049 Bytes
79e7646 a1c101c 79e7646 a1c101c 79e7646 a1c101c 79e7646 a1c101c 79e7646 a1c101c 79e7646 a1c101c 79e7646 a1c101c 79e7646 a1c101c 79e7646 a1c101c 79e7646 a1c101c 79e7646 b43fb22 79e7646 a1c101c 79e7646 a1c101c b43fb22 a1c101c 79e7646 a1c101c 79e7646 a1c101c b43fb22 a1c101c b43fb22 a1c101c b43fb22 a1c101c 79e7646 a1c101c 79e7646 a1c101c 79e7646 a1c101c |
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 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 |
import gradio as gr
from inference import generate_with_lora
from background_edit import run_background_removal_and_inpaint
import traceback, torch, gc
# βββββββββββββββββββββ Helpers βββββββββββββββββββββ
def _print_trace():
traceback.print_exc()
def safe_generate_with_lora(*a, **kw):
try:
return generate_with_lora(*a, **kw)
except gr.Error:
_print_trace()
raise
except Exception as e:
_print_trace()
raise gr.Error(f"Image generation failed: {e}")
def unload_models():
torch.cuda.empty_cache()
gc.collect()
def safe_run_background(*args, **kwargs):
try:
unload_models() # free VRAM before loading the inpainting model
return run_background_removal_and_inpaint(*args, **kwargs)
except Exception as e:
_print_trace()
# We still raise gr.Error so the wrapper will catch & log
raise gr.Error(f"[Step 2] Background replacement failed: {type(e).__name__}: {e}")
# βββββββββββββββββββββ UI βββββββββββββββββββββ
shared_output = gr.State() # holds the Step 1 output image
original_input = gr.State() # holds the original upload (optional)
with gr.Blocks() as demo:
demo.queue() # enable batching / concurrency
# βββββββββββ STEP 1: Headshot Refinement βββββββββββ
with gr.Tab("Step 1: Headshot Refinement"):
with gr.Row():
input_image = gr.Image(type="pil", label="Upload Headshot")
output_image = gr.Image(type="pil", label="Refined Output")
with gr.Row():
prompt = gr.Textbox(
label="Prompt",
value="a professional corporate headshot of a confident woman in her 30s with blonde hair"
)
negative_prompt = gr.Textbox(
label="Negative Prompt",
value="deformed, cartoon, anime, illustration, painting, drawing, sketch, low resolution, blurry, out of focus, pixelated"
)
with gr.Row():
strength = gr.Slider(0.1, 1.0, value=0.20, step=0.05, label="Strength")
guidance = gr.Slider(1, 20, value=17.0, step=0.5, label="Guidance Scale")
run_btn = gr.Button("Generate")
def _save_to_state(img):
return {"step1": img} if img is not None else gr.skip()
# Build the click-chain and store it in `event`
event = (
run_btn.click(
fn=safe_generate_with_lora,
inputs=[input_image, prompt, negative_prompt, strength, guidance],
outputs=output_image,
)
.then(_save_to_state, output_image, shared_output)
.then(lambda x: x, input_image, original_input)
)
# βββββββββββ STEP 2: Background Replacement βββββββββββ
with gr.Tab("Step 2: Replace Background"):
# This textbox will show any error messages from Step 2
error_box = gr.Textbox(label="Error", interactive=False, lines=2)
with gr.Row():
inpaint_prompt = gr.Textbox(
label="New Background Prompt",
value="modern open-plan startup office background, natural lighting, glass walls, clean design, minimalistic decor"
)
inpaint_negative = gr.Textbox(
label="Negative Prompt",
value="dark lighting, cluttered background, fantasy elements, cartoon, anime, painting, low quality, distorted shapes"
)
with gr.Row():
inpaint_result = gr.Image(type="pil", label="Inpainted Image")
with gr.Row():
inpaint_btn = gr.Button("Remove Background & Inpaint", interactive=False)
def guarded_inpaint(img, prompt_bg, neg_bg):
# If no image is in state, return an error immediately
if img is None:
return None, "Error: No headshot foundβplease run Step 1 first."
try:
print("[DEBUG] Starting background removal and inpaintingβ¦", flush=True)
result = safe_run_background(img, prompt_bg, neg_bg)
return result, "" # success: show result, clear error
except Exception as e:
# Log to console/Space logs
print(f"[Step 2 ERROR] {type(e).__name__}: {e}", flush=True)
return None, f"{type(e).__name__}: {e}"
inpaint_btn.click(
fn=guarded_inpaint,
inputs=[shared_output, inpaint_prompt, inpaint_negative],
outputs=[inpaint_result, error_box], # wire both outputs!
)
# Enable the Step 2 button once Step 1βs chain finishes
event.then(lambda: gr.update(interactive=True), None, inpaint_btn)
# Launch in debug mode so exceptions & prints go to your terminal/logs
demo.launch(debug=True)
|