app.py

import gradio as gr
import numpy as np
import random
import torch

import io, json
from PIL import Image
import os.path
from weight_fusion import compose_concepts
from regionally_controlable_sampling import sample_image, build_model, prepare_text

device = "cuda" if torch.cuda.is_available() else "cpu"
power_device = "GPU" if torch.cuda.is_available() else "CPU"

MAX_SEED = 100_000

def generate(region1_concept,
                region2_concept,
                prompt,
                pose_image_name,
                region1_prompt,
                region2_prompt,
                negative_prompt,
                region_neg_prompt,
                seed,
                randomize_seed,
                sketch_adaptor_weight,
                keypose_adaptor_weight
                ):

    if region1_concept==region2_concept:
        raise gr.Error("Please choose two different characters for merging weights.")
    if len(pose_image_name)==0:
        raise gr.Error("Please select one spatial condition!")
    if len(region1_prompt)==0 or len(region1_prompt)==0:
        raise gr.Error("Your regional prompt cannot be empty.")
    if len(prompt)==0:
        raise gr.Error("Your global prompt cannot be empty.")
    
    if randomize_seed:
        seed = random.randint(0, MAX_SEED)

    region1_concept, region2_concept = region1_concept.lower(), region2_concept.lower()
    pretrained_model = merge(region1_concept, region2_concept)

    with open('multi-concept/pose_data/pose.json') as f:
        d = json.load(f)

    pose_image = {os.path.basename(obj['img_dir']):obj for obj in d}[pose_image_name]
    # pose_image = {obj.pop('pose_id'):obj for obj in d}[int(pose_image_id)]
    print(pose_image)
    keypose_condition = pose_image['img_dir']
    region1 = pose_image['region1']
    region2 = pose_image['region2']

    region1_prompt = f'[<{region1_concept}1> <{region1_concept}2>, {region1_prompt}]'
    region2_prompt = f'[<{region2_concept}1> <{region2_concept}2>, {region2_prompt}]'
    prompt_rewrite=f"{region1_prompt}-*-{region_neg_prompt}-*-{region1}|{region2_prompt}-*-{region_neg_prompt}-*-{region2}"
    print(prompt_rewrite)

    result = infer(pretrained_model,
                  prompt,
                  prompt_rewrite,
                  negative_prompt,
                  seed,
                  keypose_condition,
                  keypose_adaptor_weight,
                #   sketch_condition,
                #   sketch_adaptor_weight,
                  )
    
    return result

def merge(concept1, concept2):
    device = "cuda" if torch.cuda.is_available() else "cpu"
    c1, c2 = sorted([concept1, concept2])
    assert c1!=c2
    merge_name = c1+'_'+c2

    save_path = f'experiments/multi-concept/{merge_name}'

    if os.path.isdir(save_path):
        print(f'{save_path} already exists. Collecting merged weights from existing weights...')

    else:
        os.makedirs(save_path)
        json_path = os.path.join(save_path,'merge_config.json')
        alpha = 1.8
        data = [
            {
                "lora_path": f"experiments/single-concept/{c1}/models/edlora_model-latest.pth",
                "unet_alpha": alpha,
                "text_encoder_alpha": alpha,
                "concept_name": f"<{c1}1> <{c1}2>"
            },
            {
                "lora_path": f"experiments/single-concept/{c2}/models/edlora_model-latest.pth",
                "unet_alpha": alpha,
                "text_encoder_alpha": alpha,
                "concept_name": f"<{c2}1> <{c2}2>"
            }
        ]
        with io.open(json_path,'w',encoding='utf8') as outfile:
            json.dump(data, outfile, indent = 4, ensure_ascii=False)

        compose_concepts(
            concept_cfg=json_path,
            optimize_textenc_iters=500,
            optimize_unet_iters=50,
            pretrained_model_path="nitrosocke/mo-di-diffusion",
            save_path=save_path,
            suffix='base',
            device=device,
        )
        print(f'Merged weight for {c1}+{c2} saved in {save_path}!\n\n')

    modelbase_path = os.path.join(save_path,'combined_model_base')
    assert os.path.isdir(modelbase_path)

    # save_path = 'experiments/multi-concept/elsa_moana_weight18/combined_model_base'
    return modelbase_path

def infer(pretrained_model,
          prompt,
          prompt_rewrite,
          negative_prompt='',
          seed=16141,
          keypose_condition=None,
          keypose_adaptor_weight=1.0,
          sketch_condition=None,
          sketch_adaptor_weight=0.0,
          region_sketch_adaptor_weight='',
          region_keypose_adaptor_weight=''
          ):

    device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
    pipe = build_model(pretrained_model, device)

    if sketch_condition is not None and os.path.exists(sketch_condition):
        sketch_condition = Image.open(sketch_condition).convert('L')
        width_sketch, height_sketch = sketch_condition.size
        print('use sketch condition')
    else:
        sketch_condition, width_sketch, height_sketch = None, 0, 0
        print('skip sketch condition')

    if keypose_condition is not None and os.path.exists(keypose_condition):
        keypose_condition = Image.open(keypose_condition).convert('RGB')
        width_pose, height_pose = keypose_condition.size
        print('use pose condition')
    else:
        keypose_condition, width_pose, height_pose = None, 0, 0
        print('skip pose condition')

    if width_sketch != 0 and width_pose != 0:
        assert width_sketch == width_pose and height_sketch == height_pose, 'conditions should be same size'
    width, height = max(width_pose, width_sketch), max(height_pose, height_sketch)
    kwargs = {
        'sketch_condition': sketch_condition,
        'keypose_condition': keypose_condition,
        'height': height,
        'width': width,
    }

    prompts = [prompt]
    prompts_rewrite = [prompt_rewrite]
    input_prompt = [prepare_text(p, p_w, height, width) for p, p_w in zip(prompts, prompts_rewrite)]
    save_prompt = input_prompt[0][0]
    print(save_prompt)

    image = sample_image(
        pipe,
        input_prompt=input_prompt,
        input_neg_prompt=[negative_prompt] * len(input_prompt),
        generator=torch.Generator(device).manual_seed(seed),
        sketch_adaptor_weight=sketch_adaptor_weight,
        region_sketch_adaptor_weight=region_sketch_adaptor_weight,
        keypose_adaptor_weight=keypose_adaptor_weight,
        region_keypose_adaptor_weight=region_keypose_adaptor_weight,
        **kwargs)
    
    return image[0]


def on_select(evt: gr.SelectData):  # SelectData is a subclass of EventData
    return evt.value['image']['orig_name']

examples_context = [
    'walking at Stanford university campus',
    'in a castle',
    'in the forest',
    'in front of Eiffel tower'
]

examples_region1 = ['wearing red hat, high resolution, best quality']
examples_region2 = ['smilling, wearing blue shirt, high resolution, best quality']

with open('multi-concept/pose_data/pose.json') as f:
    d = json.load(f)
pose_image_list = [(obj['img_id'],obj['img_dir']) for obj in d]

css="""
#col-container {
    margin: 0 auto;
    max-width: 600px;
}
"""

with gr.Blocks(css=css) as demo:
    gr.Markdown(f"""
    # Orthogonal Adaptation
    Describe your world with a **🪄 text prompt (global and local)** and choose two characters to merge. 
    Select their **👯 poses (spatial conditions)** for regionally controllable sampling to generate a unique image using our model.
    Let your creativity run wild! (Currently running on : {power_device} )    
    """)
    
    with gr.Row():
        with gr.Column(elem_id="col-container"):
            
            # gr.Markdown(f"""
            # ### 🪄 Global and Region prompts
            # """)
            # with gr.Group():   
            with gr.Tab('🪄 Global and Region prompts'):
                prompt = gr.Text(
                        label="ContextPrompt",
                        show_label=False,
                        max_lines=1,
                        placeholder="Enter your global context prompt",
                        container=False,
                    )
                

                with gr.Row():
                    
                    region1_concept = gr.Dropdown(
                        ["Elsa", "Moana"],
                        label="Character 1",
                        info="Will add more characters later!"
                    )
                    region2_concept = gr.Dropdown(
                        ["Elsa", "Moana"], 
                        label="Character 2", 
                        info="Will add more characters later!"
                    )


                with gr.Row():

                    region1_prompt = gr.Textbox(
                        label="Region1 Prompt",
                        show_label=False,
                        max_lines=2,
                        placeholder="Enter your regional prompt for character 1",
                        container=False,
                    )

                    region2_prompt = gr.Textbox(
                        label="Region2 Prompt",
                        show_label=False,
                        max_lines=2,
                        placeholder="Enter your regional prompt for character 2",
                        container=False,
                    )


                gr.Examples(
                                label = 'Global Prompt example',
                                examples = examples_context,
                                inputs = [prompt]
                            )  
                
                with gr.Row():
                    gr.Examples(
                        label = 'Region1 Prompt example',
                        examples = examples_region1,
                        inputs = [region1_prompt]
                    )

                    gr.Examples(
                        label = 'Region2 Prompt example',
                        examples = [examples_region2],
                        inputs = [region2_prompt]
                    )

            # gr.Markdown(f"""
            # ### 👯 Spatial Condition  
            # """)
            # with gr.Group():
            with gr.Tab('👯 Spatial Condition '):
                gallery = gr.Gallery(label = "Select pose for characters",
                                    value = [obj[1]for obj in pose_image_list], 
                                    elem_id = [obj[0]for obj in pose_image_list],
                                    interactive=False, show_download_button=False,
                                    preview=True, height = 400, object_fit="scale-down")
                
                pose_image_name = gr.Textbox(visible=False)
                gallery.select(on_select, None, pose_image_name)

            run_button = gr.Button("Run", scale=1)

            with gr.Accordion("Advanced Settings", open=False):

                negative_prompt = gr.Text(
                    label="Context Negative prompt",
                    max_lines=1,
                    value = 'saturated, cropped, worst quality, low quality',
                    visible=False,
                )

                region_neg_prompt =  gr.Text(
                    label="Regional Negative prompt",
                    max_lines=1,
                    value = 'shirtless, nudity, saturated, cropped, worst quality, low quality',
                    visible=False,
                )
                
                seed = gr.Slider(
                    label="Seed",
                    minimum=0,
                    maximum=MAX_SEED,
                    step=1,
                    value=0,
                )
                
                randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
                
                with gr.Row():
                    
                    sketch_adaptor_weight = gr.Slider(
                            label="Sketch Adapter Weight",
                            minimum = 0,
                            maximum = 1,
                            step=0.01,
                            value=0,
                    )
                    
                    keypose_adaptor_weight = gr.Slider(
                            label="Keypose Adapter Weight",
                            minimum = 0,
                            maximum = 1,
                            step= 0.01,
                            value=1.0,
                    )
        
        with gr.Column():
            result = gr.Image(label="Result", show_label=False)

    gr.Markdown(f"""
                *Image generation may take longer for the first time you use a new combination of characters. <br />
                This is because the model needs to load weights for each concept involved.*
                """)

    run_button.click(
        fn = generate,
        inputs = [region1_concept,
                  region2_concept,
                  prompt,
                  pose_image_name,
                  region1_prompt,
                  region2_prompt,
                  negative_prompt,
                  region_neg_prompt,
                  seed,
                  randomize_seed,
                #   sketch_condition,
                #   keypose_condition,
                  sketch_adaptor_weight,
                  keypose_adaptor_weight
                  ],
        outputs = [result]
    )

demo.queue().launch(share=True)