rembg-aws-lambda's Introduction

Rembg (AWS Lambda)

This is a stripped-down fork of danielgatis/rembg designed for AWS Lambda environments.

rembg-aws-lambda is a tool to remove images background.

Check out my similar project, profile-photo, which can create a headshot from an image.

Requirements

python: >3.7, <3.11

Installation

CPU support:

pip install rembg-aws-lambda

GPU support:

First of all, you need to check if your system supports the onnxruntime-gpu.

Go to https://onnxruntime.ai and check the installation matrix.

If yes, just run:

pip install rembg-aws-lambda[gpu]

Usage as a library

Input and output as bytes

from rembg import remove

input_path = 'input.png'
output_path = 'output.png'

with open(input_path, 'rb') as i:
    with open(output_path, 'wb') as o:
        input = i.read()
        output = remove(input)
        o.write(output)

Input and output as a PIL image

from rembg import remove
from PIL import Image

input_path = 'input.png'
output_path = 'output.png'

input = Image.open(input_path)
output = remove(input)
output.save(output_path)

Input and output as a numpy array

from rembg import remove
import cv2

input_path = 'input.png'
output_path = 'output.png'

input = cv2.imread(input_path)
output = remove(input)
cv2.imwrite(output_path, output)

How to iterate over files in a performatic way

from pathlib import Path
from rembg import remove, new_session

session = new_session()

for file in Path('path/to/folder').glob('*.png'):
    input_path = str(file)
    output_path = str(file.parent / (file.stem + ".out.png"))

    with open(input_path, 'rb') as i:
        with open(output_path, 'wb') as o:
            input = i.read()
            output = remove(input, session=session)
            o.write(output)

Models

All models are downloaded and saved in the user home folder in the .u2net directory.

The available models are:

u2net (download, source): A pre-trained model for general use cases.
u2netp (download, source): A lightweight version of u2net model.
u2net_human_seg (download, source): A pre-trained model for human segmentation.
u2net_cloth_seg (download, source): A pre-trained model for Cloths Parsing from human portrait. Here clothes are parsed into 3 category: Upper body, Lower body and Full body.
silueta (download, source): Same as u2net but the size is reduced to 43Mb.

How to train your own model

If You need more fine tunned models try this: danielgatis/rembg#193 (comment)

Some video tutorials

References

Buy me a coffee

Liked some of my work? Buy me a coffee (or more likely a beer)

License

Licensed under MIT License

rembg-aws-lambda's People

Contributors

Stargazers

Watchers

rembg-aws-lambda's Issues

alpha_matting True not work

First of all thanks for the nice tool!

i want to use remove function with alpha_matting True but when I run it, it gives me notImpemented errror.

So I replaced all commented line in rembg/bg.py and it works well.

Is there any reason that you make those code lines to comments?

Readme docs for instalation

Hey there,

are you planing to add some basic docs, how to install and run this package on lambda? For example using serverless framework?

Thank you. Good work!

[BUG] File Size Limit @PyPI

Describe the bug
When deploying to PyPI, we run into a deploy issue by default, as I bundle a model file u2net.onnx within the package and that is over 100 MB in size (actual: ~168MB).

Expected behavior
PyPI package deploys without issue to PyPI.

OS Version:
iOS 22

Rembg version:
v2.0.21

Additional context
I've opened a File Limit Increase request on the PyPI repo on GH, here:
pypi/support#2672

Edit: Looks like that got approved. The one way to test is to create a new (minor) release and test that the GH Actions workflow deploys to PyPI.

Recommend Projects

rnag / rembg-aws-lambda Goto Github PK