Skip to main content
Uber
Uber

Perception and prediction

Recent ATG R&D publications

Discrete Residual Flow for Probabilistic Pedestrian Behavior Prediction

Read more

Identifying Unknown Instances for Autonomous Driving

Read more

DeepPruner: Learning Efficient Stereo Matching via Differentiable PatchMatch

Read more

Learning Joint 2D-3D Representations for Depth Completion

Read more

End-to-End Interpretable Neural Motion Planner

Read more

UPSNet: A Unified Panoptic Segmentation Network

Read more

Multi-Task Multi-Sensor Fusion for 3D Object Detection

Read more

Deep Rigid Instance Scene Flow

Read more

DeepSignals: Predicting Intent of Drivers Through Visual Signals

Read more

Deep Multi-Sensor Lane Detection

Read more

HDNET: Exploiting HD Maps for 3D Object Detection

Read more

IntentNet: Learning to Predict Intention from Raw Sensor Data

Read more

Efficient Convolutions for Real-Time Semantic Segmentation of 3D Point Clouds

Read more

Deep Continuous Fusion for Multi-Sensor 3D Object Detection

Read more

SBNet: Sparse Blocks Network for Fast Inference

Read more

Deep Parametric Continuous Convolutional Neural Networks

Read more

PIXOR: Real-Time 3D Object Detection from Point Clouds

Read more

Matching Adversarial Networks

Read more

Fast and Furious: Real Time End-to-End 3D Detection, Tracking and Motion Forecasting with a Single Convolutional Net

Read more

End-to-End Learning of Multi-Sensor 3D Tracking by Detection

Read more

R&D Homepage

Back to Homepage