Promote Sign in Sign up

Avatar animation

Live-Portrait ✓

Animate portraits with video-driven facial movements.

Source rating 1.0 (1) 1k views 1.7M/mo traffic 11 saves Paid
Portrait Animation Driving Video Input Digital Art Creation
More categories (4)
Nvidia A40 GPU InsightFace Buffalo L Models Virtual Avatar Design Avatar animation
Visit website Sign in to save

from $0.0001

Verdict

A quick read before you dig into the full page.

Best if
Animate portraits with video-driven facial movements
Skip if
Limited to portrait animation
Pick this when
Animates portraits with video input

Considering Krikey AI instead? Compare side by side

Overview

The fofr/live-portrait is an AI model that enables portrait animation by leveraging a driving video source. The main functionality is to generate animated portraits given input images.

This happens on the platform Replicate and the API enables integration with other applications. With an open-source provision, users also have the option of running the model on their local systems with Docker.

The model, which operates on Nvidia A40 (Large) GPU hardware, varies prediction times based on the given inputs. As part of the functionality, LivePortrait uses InsightFace buffalo_l models, thus its usage in commercial applications is restricted.

Additionally, the LivePortrait code and weights are released under the MIT license. The model utilises the ComfyUI custom node that was created by Kijai and the safetensor weights they converted.

There is considerable support with documentation and available examples to assist users in understanding how to use the model. This can be especially helpful while exploring various applications of portrait animation, from digital art creation to virtual avatar design.

Live-Portrait website

Access & connections

Where you can use Live-Portrait, and what it plugs into.

Platforms

Web

Integrations

Docker

Capabilities

What this tool accepts, produces, and is strongest at.

Input

  • API
  • Video

Output

  • Video
  • API

Pros & cons

Pros

  • Animates portraits with video input
  • Integration ability with API
  • Runs on Replicate platform
  • Option for local system execution
  • Compatible with Docker
  • Operates on Nvidia GPU
  • Open-source provision
  • MIT license
Show 15 more
  • InsightFace buffalo_l model usage
  • Custom ComfyUI node
  • Assistance with documentation
  • Availability of application examples
  • Potential for digital art creation
  • Avatar design capabilities
  • Adaptable prediction times
  • Cost transparency
  • Option for public runs
  • Support for various input/output
  • Detailed README
  • Retargeting control in animation
  • Links to example driving videos
  • Links to GitHub and paper
  • Restriction for commercial applications

Cons

  • Commercial use restricted
  • Runs on Nvidia A40 GPU only
  • Variable prediction times
  • Requires Docker for local execution
  • Relies on InsightFace buffalo_l models
  • Requires ComfyUI custom node
  • Depends on safetensor weights
  • Limited to portrait animation
Show 2 more
  • MIT License restrictions
  • Usage bound to Replicate platform

Compare alternatives

Similar tools in the catalog — with a clear reason to open each one.

Why look Freemium alternative

Full comparison

Free + from $4.99/mo

Full comparison

Why look Freemium alternative

Full comparison

Why look Stronger for image work

Full comparison

Free + from $34/mo

Full comparison

from $12.99

Full comparison

Releases 1

Initial release July 8, 2024 Current
Initial release of Live-Portrait.

FAQ 12

What is the primary functionality of the LivePortrait AI model?
The primary functionality of the LivePortrait AI model is to generate animated portraits given input images.
How does LivePortrait use a driving video source to animate portraits?
LivePortrait leverages a driving video source to animate portraits. It uses the motions and expressions in the source video to animate the given portrait. Although the specific methodology isn't detailed, this generally involves feature extraction, mapping, and synthesis.
What platform does the LivePortrait operate on?
LivePortrait operates on the Replicate platform.
Is there an API for integrating LivePortrait with other applications?
Yes, there is an API provided for integrating LivePortrait with other applications.
Is it possible to run the LivePortrait model on my local system?
Yes, thanks to its open-source provision, you could run the LivePortrait model on your local system. Particularly, the Docker option they provide can assist in this.
What hardware does LivePortrait operate on?
LivePortrait operates on Nvidia A40 (Large) GPU hardware.
How do input variations affect the prediction times of the LivePortrait?
The prediction times of LivePortrait vary based on the inputs. While specific details are not mentioned, factors such as image complexity or resolution could potentially impact prediction times.
Why is the usage of LivePortrait in commercial applications restricted?
The reason commercial use of LivePortrait is restricted is because the model uses InsightFace buffalo_l models, which have their own usage limitations.
Under what license are the LivePortrait code and weights released?
The LivePortrait code and weights are released under the MIT license.
What is the ComfyUI custom node and how does LivePortrait utilize it?
The ComfyUI custom node is a component used by the LivePortrait model, created by Kijai. However, precise details of how the model utilises this node are not mentioned.
Who converted the safetensor weights the model uses and why?
The safetensor weights that the model uses were converted by Kijai, though the specific reasons for this conversion are not indicated.
What is the role of the safetensor weights in the LivePortrait's performance?
While the exact role of safetensor weights in LivePortrait's performance isn't expressly stated, they are most likely integral to the operations of the model and its ability to generate animated portraits.
Live-Portrait from $0.0001
Visit website