added resizing modes

added more info into readme
This commit is contained in:
AUTOMATIC 2022-08-24 10:52:41 +03:00
parent 2a0f8afb29
commit 1463d44faf
3 changed files with 60 additions and 3 deletions

View File

@ -92,7 +92,6 @@ If you use this feature, batch count will be ignored, because the number of pict
depends on your prompts, but batch size will still work (generating multiple pictures at the depends on your prompts, but batch size will still work (generating multiple pictures at the
same time for a small speed boost). same time for a small speed boost).
### Flagging ### Flagging
Click the Flag button under the output section, and generated images will be saved to `log/images` directory, and generation parameters Click the Flag button under the output section, and generated images will be saved to `log/images` directory, and generation parameters
will be appended to a csv file `log/log.csv` in the `/sd` directory. will be appended to a csv file `log/log.csv` in the `/sd` directory.
@ -107,3 +106,24 @@ If you generate multiple pictures, the displayed seed will be the seed of the fi
### Correct seeds for batches ### Correct seeds for batches
If you use a seed of 1000 to generate two batches of two images each, four generated images will have seeds: `1000, 1001, 1002, 1003`. If you use a seed of 1000 to generate two batches of two images each, four generated images will have seeds: `1000, 1001, 1002, 1003`.
Previous versions of the UI would produce `1000, x, 1001, x`, where x is an iamge that can't be generated by any seed. Previous versions of the UI would produce `1000, x, 1001, x`, where x is an iamge that can't be generated by any seed.
### Resizing
There are three options for resizing input images in img2img mode:
- Just resize - simply resizes source image to target resolution, resulting in incorrect aspect ratio
- Crop and resize - resize source image preserving aspect ratio so that entirety of target resolution is occupied by it, and crop parts that stick out
- Resize and fill - resize source image preserving aspect ratio so that it entirely fits target resolution, and fill empty space by rows/columns from source image
Example:
![](images/resizing.jpg)
### Loading
Gradio's loading graphic has a very negative effect on the processing speed onthe neural network.
My RTX 3090 makes images about 10% faster when the tab with gradio is not active. By defaul, the UI
now hides loading progress animation and replaces it with static "Loading..." text. Use
the --no-progressbar-hiding commandline option to revert this and show loading animations.
### Prompt validation
Stable Diffusion has a limit for imput text length. If your prompt is too long, you will get a
warning in the text output field, showing which parts of your text were truncated and consequently
ignored by the model.

BIN
images/resizing.jpg Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 453 KiB

View File

@ -241,6 +241,42 @@ def draw_prompt_matrix(im, width, height, all_prompts):
return result return result
def resize_image(resize_mode, im, width, height):
if resize_mode == 0:
res = im.resize((width, height), resample=LANCZOS)
elif resize_mode == 1:
ratio = width / height
src_ratio = im.width / im.height
src_w = width if ratio > src_ratio else im.width * height // im.height
src_h = height if ratio <= src_ratio else im.height * width // im.width
resized = im.resize((src_w, src_h), resample=LANCZOS)
res = Image.new("RGB", (width, height))
res.paste(resized, box=(width // 2 - src_w // 2, height // 2 - src_h // 2))
else:
ratio = width / height
src_ratio = im.width / im.height
src_w = width if ratio < src_ratio else im.width * height // im.height
src_h = height if ratio >= src_ratio else im.height * width // im.width
resized = im.resize((src_w, src_h), resample=LANCZOS)
res = Image.new("RGB", (width, height))
res.paste(resized, box=(width // 2 - src_w // 2, height // 2 - src_h // 2))
if ratio < src_ratio:
fill_height = height // 2 - src_h // 2
res.paste(resized.resize((width, fill_height), box=(0, 0, width, 0)), box=(0, 0))
res.paste(resized.resize((width, fill_height), box=(0, resized.height, width, resized.height)), box=(0, fill_height + src_h))
else:
fill_width = width // 2 - src_w // 2
res.paste(resized.resize((fill_width, height), box=(0, 0, 0, height)), box=(0, 0))
res.paste(resized.resize((fill_width, height), box=(resized.width, 0, resized.width, height)), box=(fill_width + src_w, 0))
return res
def check_prompt_length(prompt, comments): def check_prompt_length(prompt, comments):
"""this function tests if prompt is too long, and if so, adds a message to comments""" """this function tests if prompt is too long, and if so, adds a message to comments"""
@ -488,7 +524,7 @@ txt2img_interface = gr.Interface(
) )
def img2img(prompt: str, init_img, ddim_steps: int, use_GFPGAN: bool, prompt_matrix, n_iter: int, batch_size: int, cfg_scale: float, denoising_strength: float, seed: int, height: int, width: int): def img2img(prompt: str, init_img, ddim_steps: int, use_GFPGAN: bool, prompt_matrix, n_iter: int, batch_size: int, cfg_scale: float, denoising_strength: float, seed: int, height: int, width: int, resize_mode: int):
outpath = opt.outdir or "outputs/img2img-samples" outpath = opt.outdir or "outputs/img2img-samples"
sampler = KDiffusionSampler(model) sampler = KDiffusionSampler(model)
@ -498,7 +534,7 @@ def img2img(prompt: str, init_img, ddim_steps: int, use_GFPGAN: bool, prompt_mat
def init(): def init():
image = init_img.convert("RGB") image = init_img.convert("RGB")
image = image.resize((width, height), resample=LANCZOS) image = resize_image(resize_mode, image, width, height)
image = np.array(image).astype(np.float32) / 255.0 image = np.array(image).astype(np.float32) / 255.0
image = image[None].transpose(0, 3, 1, 2) image = image[None].transpose(0, 3, 1, 2)
image = torch.from_numpy(image) image = torch.from_numpy(image)
@ -562,6 +598,7 @@ img2img_interface = gr.Interface(
gr.Number(label='Seed', value=-1), gr.Number(label='Seed', value=-1),
gr.Slider(minimum=64, maximum=2048, step=64, label="Height", value=512), gr.Slider(minimum=64, maximum=2048, step=64, label="Height", value=512),
gr.Slider(minimum=64, maximum=2048, step=64, label="Width", value=512), gr.Slider(minimum=64, maximum=2048, step=64, label="Width", value=512),
gr.Radio(label="Resize mode", choices=["Just resize", "Crop and resize", "Resize and fill"], type="index", value="Just resize")
], ],
outputs=[ outputs=[
gr.Gallery(), gr.Gallery(),