Google's Objectron model can track 3D objects from 2D imagery, and it's lightweight enough to run on any modern smartphone.

For multi-object tracking, IMOU optimizes object detection, re-identification, object correlation, and other modules based on the TBD framework to improve the overall performance.

Nvidia Research and others collaborated to create the DIB-R framework that can predict 3D properties from 2D images to create 3D models.

They say online media is going to be 3D, starting with e-tailing. Where can a merchant get that multi-dimensional content? They have 2D pictures, and they have physical objects. Occasionally there ...

NVIDIA has developed a new NeRF technique — the fastest one to date, the company claims — that only needs seconds to train and to generate a 3D scene.

The quality improves but the computational cost skyrockets. That’s held back the otherwise highly desirable possibility of extrapolating 3D forms from 2D images.

A team has created an application called PartCrafter that creates 3D models, in single or multiple parts, using AI based on 2D images for input.

Recently, IMOU, the smart home brand in China, wins the first places in KITTI 2D object detection of pedestrian, multi-object tracking of pedestrian and car evaluations.