Apple Research Revealed

In what makes one wonder the intent behind this move, two Apple scientists have published a paper revealing their research into improving real time object recognition. One of them is an AI researcher and the other specializes in machine learning. They are working on accurate detection of objects including those derived from what they refer to as sparse lidar 3D point clouds.

They are using a novel, voxel – based approach that they describe as “removing the need for manual feature engineering by using VoxelNet, a generic 3D detection network that unifies feature extraction and bounding box prediction into a single stage, end-to-end trainable deep network.”

From the paper, “Our approach can operate directly on sparse 3D points and capture 3D shape information effectively. We also present an efficient implementation of VoxelNet that benefits from point cloud sparsity and parallel processing on a voxel grid.”

The robotics community has been working on this problem for 20+ years. It will be interesting to see if the autonomous vehicle funding can deliver on the performance needed to support highway speed automation.

You can read the full paper here.

This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *

*