Please enable Javascript
Skip to main content

Core AI tech

Recent ATG R&D publications

Hierarchical Verification for Adversarial Robustness

Conditional Entropy Coding for Efficient Video Compression

Learning Lane Graph Representations for Motion Forecasting

LevelSet R-CNN: A Deep Variational Method for Instance Segmentation

OctSqueeze: Octree-Structured Entropy Model for LiDAR Compression

DSIC: Deep Stereo Image Compression

Graph HyperNetworks for Neural Architecture Search

SBNet: Sparse Blocks Network for Fast Inference

Deep Parametric Continuous Convolutional Neural Networks

Matching Adversarial Networks

Learning to Reweight Examples for Robust Deep Learning

R&D Homepage