app.py

import gradio as gr
import numpy as np
import torch
import jax
import jax.numpy as jnp
from diffusers import StableDiffusionInpaintPipeline
from flax.jax_utils import replicate
from flax.training.common_utils import shard
from PIL import Image
from segment_anything import SamPredictor, sam_model_registry, SamAutomaticMaskGenerator
from diffusers import (
    UniPCMultistepScheduler,
    FlaxStableDiffusionControlNetPipeline,
    FlaxControlNetModel,
)

import colorsys

sam_checkpoint = "sam_vit_h_4b8939.pth"
model_type = "vit_h"
device = "cpu"


sam = sam_model_registry[model_type](checkpoint=sam_checkpoint)
sam.to(device=device)
predictor = SamPredictor(sam)
mask_generator = SamAutomaticMaskGenerator(sam)


controlnet, controlnet_params = FlaxControlNetModel.from_pretrained(
    "mfidabel/controlnet-segment-anything", dtype=jnp.float32
)

pipe, params = FlaxStableDiffusionControlNetPipeline.from_pretrained(
    "runwayml/stable-diffusion-v1-5",
    controlnet=controlnet,
    revision="flax",
    dtype=jnp.bfloat16,
)

params["controlnet"] = controlnet_params
p_params = replicate(params)

pipe.scheduler = UniPCMultistepScheduler.from_config(pipe.scheduler.config)
pipe = pipe.to(device)


with gr.Blocks() as demo:
    gr.Markdown("# WildSynth: Synthetic Wildlife Data Generation")
    gr.Markdown(
        """
        We have trained a JAX ControlNet model with
    To try the demo, upload an image and select object(s) you want to inpaint.
    Write a prompt & a negative prompt to control the inpainting.
    Click on the "Submit" button to inpaint the selected object(s).
    Check "Background" to inpaint the background instead of the selected object(s).

    If the demo is slow, clone the space to your own HF account and run on a GPU.
    """
    )
    with gr.Row():
        input_img = gr.Image(label="Input")
        mask_img = gr.Image(label="Mask", interactive=False)
        output_img = gr.Image(label="Output", interactive=False)

    with gr.Row():
        prompt_text = gr.Textbox(lines=1, label="Prompt")
        negative_prompt_text = gr.Textbox(lines=1, label="Negative Prompt")

    with gr.Row():
        submit = gr.Button("Submit")
        clear = gr.Button("Clear")

    def generate_mask(image, evt: gr.SelectData):
        predictor.set_image(image)
        input_point = np.array([120, 21])
        input_label = np.ones(input_point.shape[0])
        mask, _, _ = predictor.predict(
            point_coords=input_point,
            point_labels=input_label,
            multimask_output=False,
        )

        # clear torch cache
        torch.cuda.empty_cache()
        mask = Image.fromarray(mask[0, :, :])
        segs = mask_generator.generate(image)
        boolean_masks = [s["segmentation"] for s in segs]
        finseg = np.zeros(
            (boolean_masks[0].shape[0], boolean_masks[0].shape[1], 3), dtype=np.uint8
        )
        # Loop over the boolean masks and assign a unique color to each class
        for class_id, boolean_mask in enumerate(boolean_masks):
            hue = class_id * 1.0 / len(boolean_masks)
            rgb = tuple(int(i * 255) for i in colorsys.hsv_to_rgb(hue, 1, 1))
            rgb_mask = np.zeros(
                (boolean_mask.shape[0], boolean_mask.shape[1], 3), dtype=np.uint8
            )
            rgb_mask[:, :, 0] = boolean_mask * rgb[0]
            rgb_mask[:, :, 1] = boolean_mask * rgb[1]
            rgb_mask[:, :, 2] = boolean_mask * rgb[2]
            finseg += rgb_mask

        torch.cuda.empty_cache()

        return mask, finseg

    def infer(
        image, prompts, negative_prompts, num_inference_steps=50, seed=4, num_samples=4
    ):
        try:
            rng = jax.random.PRNGKey(int(seed))
            num_inference_steps = int(num_inference_steps)
            image = Image.fromarray(image, mode="RGB")
            num_samples = max(jax.device_count(), int(num_samples))
            p_rng = jax.random.split(rng, jax.device_count())

            prompt_ids = pipe.prepare_text_inputs([prompts] * num_samples)
            negative_prompt_ids = pipe.prepare_text_inputs(
                [negative_prompts] * num_samples
            )
            processed_image = pipe.prepare_image_inputs([image] * num_samples)

            prompt_ids = shard(prompt_ids)
            negative_prompt_ids = shard(negative_prompt_ids)
            processed_image = shard(processed_image)

            output = pipe(
                prompt_ids=prompt_ids,
                image=processed_image,
                params=p_params,
                prng_seed=p_rng,
                num_inference_steps=num_inference_steps,
                neg_prompt_ids=negative_prompt_ids,
                jit=True,
            ).images

            del negative_prompt_ids
            del processed_image
            del prompt_ids

            output = output.reshape((num_samples,) + output.shape[-3:])
            final_image = [np.array(x * 255, dtype=np.uint8) for x in output]
            print(output.shape)
            del output

        except Exception as e:
            print("Error: " + str(e))
            final_image = [np.zeros((512, 512, 3), dtype=np.uint8)] * num_samples
        finally:
            gc.collect()
            return final_image

    def _clear(sel_pix, img, mask, seg, out, prompt, neg_prompt, bg):
        img = None
        mask = None
        seg = None
        out = None
        prompt = ""
        neg_prompt = ""
        bg = False
        return img, mask, seg, out, prompt, neg_prompt, bg

    input_img.change(
        generate_mask,
        inputs=[input_img],
        outputs=[mask_img],
    )
    submit.click(
        infer,
        inputs=[mask_img, prompt_text, negative_prompt_text],
        outputs=[output_img],
    )
    clear.click(
        _clear,
        inputs=[
            input_img,
            mask_img,
            output_img,
            prompt_text,
            negative_prompt_text,
        ],
        outputs=[
            input_img,
            mask_img,
            output_img,
            prompt_text,
            negative_prompt_text,
        ],
    )

if __name__ == "__main__":
    demo.queue()
    demo.launch()