OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Neural Style Pt

Neural Style Pt is a Python and PyTorch implementation of neural style transfer that uses convolutional neural networks to combine the content of one image with the artistic style of another.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Neural Style Pt — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want to generate an image that preserves content from one input while transferring visual features from one or more style images.

Who it is for

  • People creating stylized versions of photographs or other images.
  • Developers or researchers who want a command-line implementation of neural style transfer in Python and PyTorch.

Documented uses

  • Apply an artwork's visual style to a photograph.
  • Blend multiple style images and set their relative weights.
  • Adjust the content-versus-style balance, style scale and generated image size.
Before you choose

Requirements & limits

The code is publicly available under the MIT license.

  • PyTorch is required; pretrained model files must be downloaded with the included script.
  • The README warns that default settings can require substantial GPU memory and documents ways to reduce it.
  • The tool supports CPU mode and optional compute backends, with backend-specific dependencies.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up Neural Style Pt and its model files

Prepare the documented dependencies and download the pretrained models used by the tool.

Before you start

  • Python
  • PyTorch

Steps

  1. Install PyTorch.

  2. Run the command below to download the VGG-19 and VGG-16 models; VGG-19 is used by default.

    Documented command
    python models/download_models.py

Expected outcome

The model files are downloaded for use by Neural Style Pt.

Guide limitations

  • CUDA, cuDNN, ROCm, MKL and OpenMP are listed as optional backend dependencies.
  • The repository links to separate installation instructions for Ubuntu and Windows.
Documentation 1 ↗
Source-based practical guide

Generate an image using content and style inputs

Run the documented basic command to combine a content image with a style image.

Before you start

  • PyTorch and the model files downloaded using the repository setup instructions.
  • A content image and a style image.

Steps

  1. Provide paths to the content and style images.

  2. Run the command below.

    Documented command
    python neural_style.py -style_image -content_image

Expected outcome

The script generates a stylized image; the README's basic command does not specify a custom output path.

Guide limitations

  • Image paths should use relative or full absolute paths rather than ~ for the home directory.
  • The README lists a default maximum image side length of 512 pixels and 1,000 optimization iterations.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
6 November 2026
Original website title
ProGamerGov/neural-style-pt · Official homepage title · retained in its source language
How we review sourcesSuggest a correction