Add files

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.jpg filter=lfs diff=lfs merge=lfs -text

.pre-commit-config.yaml

@@ -0,0 +1,50 @@
+repos:
+- repo: https://github.com/pre-commit/pre-commit-hooks
+  rev: v4.4.0
+  hooks:
+  - id: check-executables-have-shebangs
+  - id: check-json
+  - id: check-merge-conflict
+  - id: check-shebang-scripts-are-executable
+  - id: check-toml
+  - id: check-yaml
+  - id: end-of-file-fixer
+  - id: mixed-line-ending
+    args: ["--fix=lf"]
+  - id: requirements-txt-fixer
+  - id: trailing-whitespace
+- repo: https://github.com/myint/docformatter
+  rev: v1.7.5
+  hooks:
+  - id: docformatter
+    args: ["--in-place"]
+- repo: https://github.com/pycqa/isort
+  rev: 5.12.0
+  hooks:
+    - id: isort
+      args: ["--profile", "black"]
+- repo: https://github.com/pre-commit/mirrors-mypy
+  rev: v1.5.1
+  hooks:
+    - id: mypy
+      args: ["--ignore-missing-imports"]
+      additional_dependencies: ["types-python-slugify", "types-requests", "types-PyYAML"]
+- repo: https://github.com/psf/black
+  rev: 23.9.1
+  hooks:
+    - id: black
+      language_version: python3.10
+      args: ["--line-length", "119"]
+- repo: https://github.com/kynan/nbstripout
+  rev: 0.6.1
+  hooks:
+    - id: nbstripout
+      args: ["--extra-keys", "metadata.interpreter metadata.kernelspec cell.metadata.pycharm"]
+- repo: https://github.com/nbQA-dev/nbQA
+  rev: 1.7.0
+  hooks:
+    - id: nbqa-black
+    - id: nbqa-pyupgrade
+      args: ["--py37-plus"]
+    - id: nbqa-isort
+      args: ["--float-to-top"]

.vscode/settings.json

@@ -0,0 +1,21 @@
+{
+    "[python]": {
+        "editor.defaultFormatter": "ms-python.black-formatter",
+        "editor.formatOnType": true,
+        "editor.codeActionsOnSave": {
+            "source.organizeImports": true
+        }
+    },
+    "black-formatter.args": [
+        "--line-length=119"
+    ],
+    "isort.args": ["--profile", "black"],
+    "flake8.args": [
+        "--max-line-length=119"
+    ],
+    "ruff.args": [
+        "--line-length=119"
+    ],
+    "editor.formatOnSave": true,
+    "files.insertFinalNewline": true
+}

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2023 hysts
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

README.md

@@ -1,5 +1,5 @@
 ---
-title: BLIP Diffusion
+title: BLIP-Diffusion
 emoji: 🔥
 colorFrom: yellow
 colorTo: indigo
@@ -7,6 +7,7 @@ sdk: gradio
 sdk_version: 3.44.4
 app_file: app.py
 pinned: false
+suggested_hardware: t4-small
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,30 @@
+#!/usr/bin/env python
+
+import os
+
+import gradio as gr
+import torch
+
+from app_stylization import demo as demo_stylization
+from app_zero_shot import demo as demo_zero_shot
+
+DESCRIPTION = "# [BLIP-Diffusion](https://github.com/salesforce/LAVIS/tree/main/projects/blip-diffusion)"
+
+if not torch.cuda.is_available():
+    DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"
+
+with gr.Blocks(css="style.css") as demo:
+    gr.Markdown(DESCRIPTION)
+    gr.DuplicateButton(
+        value="Duplicate Space for private use",
+        elem_id="duplicate-button",
+        visible=os.getenv("SHOW_DUPLICATE_BUTTON") == "1",
+    )
+    with gr.Tabs():
+        with gr.Tab(label="Zero-shot Subject-driven Generation"):
+            demo_zero_shot.render()
+        with gr.Tab(label="Stylization"):
+            demo_stylization.render()
+
+if __name__ == "__main__":
+    demo.queue(max_size=20).launch()

app_stylization.py

@@ -0,0 +1,185 @@
+#!/usr/bin/env python
+
+import gradio as gr
+import PIL.Image
+import spaces
+import torch
+from controlnet_aux import CannyDetector
+from diffusers.pipelines import BlipDiffusionControlNetPipeline
+
+from settings import CACHE_EXAMPLES, DEFAULT_NEGATIVE_PROMPT, MAX_INFERENCE_STEPS
+from utils import MAX_SEED, randomize_seed_fn
+
+canny_detector = CannyDetector()
+device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+if torch.cuda.is_available():
+    pipe = BlipDiffusionControlNetPipeline.from_pretrained(
+        "Salesforce/blipdiffusion-controlnet", torch_dtype=torch.float16
+    ).to(device)
+else:
+    pipe = None
+
+
+@spaces.GPU
+def run(
+    condition_image: PIL.Image.Image,
+    style_image: PIL.Image.Image,
+    condition_subject: str,
+    style_subject: str,
+    prompt: str,
+    negative_prompt: str = DEFAULT_NEGATIVE_PROMPT,
+    seed: int = 0,
+    guidance_scale: float = 7.5,
+    num_inference_steps: int = 25,
+) -> PIL.Image.Image:
+    if num_inference_steps > MAX_INFERENCE_STEPS:
+        raise gr.Error(f"Number of inference steps must be less than {MAX_INFERENCE_STEPS}")
+    condition_image = canny_detector(condition_image, 30, 70, output_type="pil")
+    return pipe(
+        prompt,
+        style_image,
+        condition_image,
+        style_subject,
+        condition_subject,
+        generator=torch.Generator(device=device).manual_seed(seed),
+        guidance_scale=guidance_scale,
+        num_inference_steps=num_inference_steps,
+        neg_prompt=negative_prompt,
+        height=512,
+        width=512,
+    ).images[0]
+
+
+with gr.Blocks() as demo:
+    with gr.Row():
+        with gr.Column():
+            with gr.Box():
+                condition_image = gr.Image(label="Condition Image")
+                style_image = gr.Image(label="Style Image")
+                condition_subject = gr.Textbox(label="Condition Subject")
+                style_subject = gr.Textbox(label="Style Subject")
+                prompt = gr.Textbox(label="Prompt")
+                run_button = gr.Button()
+            with gr.Accordion(label="Advanced options", open=False):
+                negative_prompt = gr.Textbox(label="Negative Prompt", value=DEFAULT_NEGATIVE_PROMPT)
+                seed = gr.Slider(
+                    label="Seed",
+                    minimum=0,
+                    maximum=MAX_SEED,
+                    step=1,
+                    value=0,
+                )
+                randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
+                guidance_scale = gr.Slider(
+                    label="Guidance Scale",
+                    minimum=0,
+                    maximum=10,
+                    step=0.1,
+                    value=7.5,
+                )
+                num_inference_steps = gr.Slider(
+                    label="Number of inference steps",
+                    minimum=1,
+                    maximum=MAX_INFERENCE_STEPS,
+                    step=1,
+                    value=25,
+                )
+        with gr.Column():
+            result = gr.Image(label="Result")
+
+    gr.Examples(
+        examples=[
+            [
+                "images/kettle.jpg",
+                "images/flower.jpg",
+                "teapot",
+                "flower",
+                "on a marble table",
+            ],
+        ],
+        inputs=[
+            condition_image,
+            style_image,
+            condition_subject,
+            style_subject,
+            prompt,
+        ],
+        outputs=result,
+        fn=run,
+        cache_examples=CACHE_EXAMPLES,
+    )
+
+    inputs = [
+        condition_image,
+        style_image,
+        condition_subject,
+        style_subject,
+        prompt,
+        negative_prompt,
+        seed,
+        guidance_scale,
+        num_inference_steps,
+    ]
+    condition_subject.submit(
+        fn=randomize_seed_fn,
+        inputs=[seed, randomize_seed],
+        outputs=seed,
+        queue=False,
+        api_name=False,
+    ).then(
+        fn=run,
+        inputs=inputs,
+        outputs=result,
+        api_name=False,
+    )
+    style_subject.submit(
+        fn=randomize_seed_fn,
+        inputs=[seed, randomize_seed],
+        outputs=seed,
+        queue=False,
+        api_name=False,
+    ).then(
+        fn=run,
+        inputs=inputs,
+        outputs=result,
+        api_name=False,
+    )
+    prompt.submit(
+        fn=randomize_seed_fn,
+        inputs=[seed, randomize_seed],
+        outputs=seed,
+        queue=False,
+        api_name=False,
+    ).then(
+        fn=run,
+        inputs=inputs,
+        outputs=result,
+        api_name=False,
+    )
+    negative_prompt.submit(
+        fn=randomize_seed_fn,
+        inputs=[seed, randomize_seed],
+        outputs=seed,
+        queue=False,
+        api_name=False,
+    ).then(
+        fn=run,
+        inputs=inputs,
+        outputs=result,
+        api_name=False,
+    )
+    run_button.click(
+        fn=randomize_seed_fn,
+        inputs=[seed, randomize_seed],
+        outputs=seed,
+        queue=False,
+        api_name=False,
+    ).then(
+        fn=run,
+        inputs=inputs,
+        outputs=result,
+        api_name="run-stylization",
+    )
+
+if __name__ == "__main__":
+    demo.queue(max_size=20).launch()

app_zero_shot.py

@@ -0,0 +1,174 @@
+#!/usr/bin/env python
+
+import gradio as gr
+import PIL.Image
+import spaces
+import torch
+from diffusers.pipelines import BlipDiffusionPipeline
+
+from settings import CACHE_EXAMPLES, DEFAULT_NEGATIVE_PROMPT, MAX_INFERENCE_STEPS
+from utils import MAX_SEED, randomize_seed_fn
+
+device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+if torch.cuda.is_available():
+    pipe = BlipDiffusionPipeline.from_pretrained("Salesforce/blipdiffusion", torch_dtype=torch.float16).to(device)
+else:
+    pipe = None
+
+
+@spaces.GPU
+def run(
+    condition_image: PIL.Image.Image,
+    condition_subject: str,
+    target_subject: str,
+    prompt: str,
+    negative_prompt: str = DEFAULT_NEGATIVE_PROMPT,
+    seed: int = 0,
+    guidance_scale: float = 7.5,
+    num_inference_steps: int = 25,
+) -> PIL.Image.Image:
+    if num_inference_steps > MAX_INFERENCE_STEPS:
+        raise gr.Error(f"Number of inference steps must be less than {MAX_INFERENCE_STEPS}")
+    return pipe(
+        prompt,
+        condition_image,
+        condition_subject,
+        target_subject,
+        generator=torch.Generator(device=device).manual_seed(seed),
+        guidance_scale=guidance_scale,
+        num_inference_steps=num_inference_steps,
+        neg_prompt=negative_prompt,
+        height=512,
+        width=512,
+    ).images[0]
+
+
+with gr.Blocks() as demo:
+    with gr.Row():
+        with gr.Column():
+            with gr.Box():
+                condition_image = gr.Image(label="Condition Image")
+                condition_subject = gr.Textbox(label="Condition Subject")
+                target_subject = gr.Textbox(label="Target Subject")
+                prompt = gr.Textbox(label="Prompt")
+                run_button = gr.Button()
+            with gr.Accordion(label="Advanced options", open=False):
+                negative_prompt = gr.Textbox(label="Negative Prompt", value=DEFAULT_NEGATIVE_PROMPT)
+                seed = gr.Slider(
+                    label="Seed",
+                    minimum=0,
+                    maximum=MAX_SEED,
+                    step=1,
+                    value=0,
+                )
+                randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
+                guidance_scale = gr.Slider(
+                    label="Guidance Scale",
+                    minimum=0,
+                    maximum=10,
+                    step=0.1,
+                    value=7.5,
+                )
+                num_inference_steps = gr.Slider(
+                    label="Number of inference steps",
+                    minimum=1,
+                    maximum=MAX_INFERENCE_STEPS,
+                    step=1,
+                    value=25,
+                )
+        with gr.Column():
+            result = gr.Image(label="Result")
+
+    gr.Examples(
+        examples=[
+            [
+                "images/dog.png",
+                "dog",
+                "dog",
+                "swimming underwater",
+            ],
+        ],
+        inputs=[
+            condition_image,
+            condition_subject,
+            target_subject,
+            prompt,
+        ],
+        outputs=result,
+        fn=run,
+        cache_examples=CACHE_EXAMPLES,
+    )
+
+    inputs = [
+        condition_image,
+        condition_subject,
+        target_subject,
+        prompt,
+        negative_prompt,
+        seed,
+        guidance_scale,
+        num_inference_steps,
+    ]
+    condition_subject.submit(
+        fn=randomize_seed_fn,
+        inputs=[seed, randomize_seed],
+        outputs=seed,
+        queue=False,
+        api_name=False,
+    ).then(
+        fn=run,
+        inputs=inputs,
+        outputs=result,
+        api_name=False,
+    )
+    target_subject.submit(
+        fn=randomize_seed_fn,
+        inputs=[seed, randomize_seed],
+        outputs=seed,
+        queue=False,
+        api_name=False,
+    ).then(
+        fn=run,
+        inputs=inputs,
+        outputs=result,
+        api_name=False,
+    )
+    prompt.submit(
+        fn=randomize_seed_fn,
+        inputs=[seed, randomize_seed],
+        outputs=seed,
+        queue=False,
+        api_name=False,
+    ).then(
+        fn=run,
+        inputs=inputs,
+        outputs=result,
+        api_name=False,
+    )
+    negative_prompt.submit(
+        fn=randomize_seed_fn,
+        inputs=[seed, randomize_seed],
+        outputs=seed,
+        queue=False,
+        api_name=False,
+    ).then(
+        fn=run,
+        inputs=inputs,
+        outputs=result,
+        api_name=False,
+    )
+    run_button.click(
+        fn=randomize_seed_fn,
+        inputs=[seed, randomize_seed],
+        outputs=seed,
+        queue=False,
+        api_name=False,
+    ).then(
+        fn=run,
+        inputs=inputs,
+        outputs=result,
+        api_name="run-zero-shot",
+    )
+
+if __name__ == "__main__":
+    demo.queue(max_size=20).launch()

images/LICENSE.txt

@@ -0,0 +1,14 @@
+BSD 3-Clause License
+
+Copyright (c) 2022 Salesforce, Inc.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
+
+1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
+
+2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
+
+3. Neither the name of Salesforce.com nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

images/README.md

@@ -0,0 +1,4 @@
+# About these images
+
+These images were obtained from [this link](https://github.com/salesforce/LAVIS/tree/e4040b13d6120062829ee9625f016f3cd3dd16e6/projects/blip-diffusion/images).
+The original repository is licensed under the BSD-3-Clause license.

requirements.txt

@@ -0,0 +1,8 @@
+controlnet_aux==0.0.7
+git+https://github.com/huggingface/diffusers@5b11c5d
+gradio==3.44.4
+Pillow==10.0.1
+spaces==0.14.0
+torch==2.0.0
+torchvision==0.15.1
+transformers==4.33.2

settings.py

@@ -0,0 +1,5 @@
+import os
+
+MAX_INFERENCE_STEPS = 50
+DEFAULT_NEGATIVE_PROMPT = "over-exposure, under-exposure, saturated, duplicate, out of frame, lowres, cropped, worst quality, low quality, jpeg artifacts, morbid, mutilated, out of frame, ugly, bad anatomy, bad proportions, deformed, blurry, duplicate"
+CACHE_EXAMPLES = os.getenv("CACHE_EXAMPLES") == "1"

style.css

@@ -0,0 +1,10 @@
+h1 {
+  text-align: center;
+}
+
+#duplicate-button {
+  margin: auto;
+  color: #fff;
+  background: #1565c0;
+  border-radius: 100vh;
+}

utils.py

@@ -0,0 +1,11 @@
+import random
+
+import numpy as np
+
+MAX_SEED = np.iinfo(np.int32).max
+
+
+def randomize_seed_fn(seed: int, randomize_seed: bool) -> int:
+    if randomize_seed:
+        seed = random.randint(0, MAX_SEED)
+    return seed