The Checkpoint #13 – The week in AI art

What's new

Local relighting of real scenes – Using GANs to turn on and off light sources in images. Paper, Code.

Latent Image Animator – Animating still images via latent space exploration. Github | Paper | Replicate demo 

How normal am I? – Explore how AI judges your face.

Text2LIVE

Text2LIVE: Text-driven layered image and video editing – A model that can edit objects and add effects by overlaying semi-transparent layers onto images and video.

Pulsar+Clip – Text-to-point cloud. Generates point cloud images from text with Clip.

nev @apeoffire

Introducing Pulsar+CLIP: the smaller cousin of text2voxels Instead of voxels, this notebook generates images from text with point clouds Prompt: a blue cat https://t.co/zrrxRkMRp2

Featured notebook – FILM: Frame Interpolation for Large Motion

Transforms near-duplicate images into slow motion video. While it can convert videos or still images, it will also work with just two frames, producing interesting morphing effects.

Get it:

• Github

• Colab

Thanks for reading!

If you have anything you’d like to be featured or want to get in touch, give me a shout on Twitter or via email. Please also consider supporting me on Patreon so that I can spend more time creating content like this.