Revolutionary AI tech captures and edits real objects in 3D.

TLDR:

• New AI technology called Proximity Attention Point Rendering (PAPR) allows for the transformation of 2D photos into editable 3D models.
• The technology creates a cloud of 3D points from a series of 2D photos, enabling intuitive shape editing of objects.

A new advancement in AI technology by researchers at Simon Fraser University (SFU) in Canada has introduced a technique called Proximity Attention Point Rendering (PAPR) that allows for the transformation of 2D photos into editable 3D models. This technology was showcased at the 2023 Conference on Neural Information Processing Systems (NeurIPS) and has the potential to revolutionize the way we capture and edit real-life objects in 3D.

The PAPR technology takes a series of 2D photos of an object and converts them into a cloud of 3D points, representing the object’s shape and appearance. This allows users to manipulate the object by adjusting individual points and altering its shape and appearance. The rendered 3D point cloud can be viewed from various angles and converted back into 2D photos, providing a realistic view of the edited object.

Dr. Ke Li, the senior author of the paper and an assistant professor of computer science at SFU, emphasizes the impact of AI and machine learning on reshaping the process of reconstructing 3D objects from 2D images. This breakthrough in AI technology opens up new possibilities for consumer technology, with 3D potentially becoming as common a medium for visual communication and expression as 2D is today.

The findings of the study were published in arXiv, showcasing the potential of PAPR technology to revolutionize the way we capture and edit real-life objects in 3D.