simple-lama-inpainting

What it is and what it does

simple-lama-inpainting wraps the LaMa (Large Mask Inpainting) model in a lightweight Python package, letting you fill masked regions of images using deep learning. It accepts images and binary masks as input and outputs inpainted results. The package provides both a command-line interface for batch processing and a Python API for integration into larger workflows.

The package depends on torch, torchvision, opencv-python, pillow, numpy, and fire. It was last released on 2023-07-28 and is now abandoned, meaning no active maintenance or updates to handle breaking changes in its dependencies. The underlying LaMa model is from a 2021 research paper on resolution-robust inpainting.

Use it for:

• Remove unwanted objects from photographs by masking them and running inpainting.
• Fill in damaged or missing regions of images for restoration work.
• Generate clean backgrounds by inpainting over foreground subjects in batch workflows.
• Preprocess training data by removing distracting elements from images.
• Command-line batch processing of multiple images with consistent mask-based inpainting.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Performs image inpainting using the LaMa model, filling in masked regions of images via a command-line tool or Python API.

No. The package is abandoned and has no declared license, creating both maintenance and legal risk. While it has low install friction and no known vulnerabilities, the lack of updates since 2023-07-28 means it will likely break as its dependencies evolve. Consider a maintained alternative or the original LaMa implementation directly.

Install

simple-lama-inpainting on PyPI

pip install simple-lama-inpainting

uv add simple-lama-inpainting

poetry add simple-lama-inpainting

Installing simple-lama-inpainting

Before you install

Low install friction with a pure Python wheel, but the package is abandoned as of 1113 days since its last release. Six runtime dependencies including torch and torchvision will add significant download and disk overhead.

License in practice

License treatment is unclear; the package declares no SPDX identifier or raw license text. Verify licensing before use in proprietary or commercial contexts.

Quickstart

pip install simple-lama-inpainting

from simple_lama_inpainting import SimpleLama

simple_lama = SimpleLama()
result = simple_lama(image, mask)
result.save("inpainted.png")

Requires Python >=3.10,<4.0. torch and torchvision dependencies may require a compatible CUDA toolkit or CPU-only build depending on your environment.

Verify before relying

• Whether the package works reliably with current versions of torch, torchvision, and other dependencies given its abandoned status.
• Actual license terms and attribution requirements for the underlying LaMa model and this wrapper.
• Performance characteristics and memory requirements for typical image sizes.
• Input/output format details beyond what the description excerpt specifies.

Package facts

Evidence: simple_lama_inpainting-0.1.2-py3-none-any.whl

Tags

More Artificial Intelligence packages