OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Tp Gan

TP Gan is a Python and TensorFlow implementation of TP-GAN, a generative model that synthesizes a frontal-view face image from a single image captured at another pose. Its repository also includes Matlab preprocessing for face crops aligned with five facial keypoints.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Tp Gan — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when investigating or prototyping research workflows that convert non-frontal face images into frontal views and can accommodate a separate keypoint-based preprocessing stage.

Who it is for

  • Computer-vision researchers studying face frontalization and pose normalization.
  • Developers assessing a research model for face-image synthesis pipelines or demos.

Documented uses

  • Synthesize a frontal-view image from a single angled or profile face image.
  • Prepare cropped face inputs and transformed keypoints using the provided five-keypoint alignment script.
  • Review example synthesized images and MultiPIE keypoint resources for research workflows.
Before you choose

Requirements & limits

The code is offered for free non-commercial use subject to its license. Commercial inquiries are directed to the authors.

  • The implementation was initially written for TensorFlow 0.12.
  • Input preparation uses a Matlab script and five facial keypoints.
  • The repository describes the code as an initial release that may not be fully tested.
  • Original MultiPIE data is not redistributed by the repository; some synthesized test images and keypoint files are linked.

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

Prepare a cropped face input with five keypoints

Create a cropped face image and transformed keypoints for use as TP-GAN input.

Before you start

  • A face image and five facial keypoints.
  • MATLAB and the repository script face_db_align_single_custom.m.

Steps

  1. Obtain five keypoints for the face; the project notes these can come from a facial landmark detector or manual labeling.

  2. Use face_db_align_single_custom.m to crop the input image using the five keypoints.

  3. Use the script’s outputs: the cropped image and transformed keypoints.

Expected outcome

A cropped face image and corresponding transformed keypoints, matching the preprocessing described by the project.

Guide limitations

  • The repository describes the code as an initial release that may not be fully tested.
  • The README does not provide further invocation instructions for the MATLAB script.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
18 October 2026
Original website title
HRLTY/TP-GAN · Official homepage title · retained in its source language
How we review sourcesSuggest a correction