Please enable Javascript
Пропустить и перейти к основному содержимому

Perception and prediction

Recent ATG R&D publications

DSDNet: Deep Structured self-Driving Network

Importance of Prior Knowledge in Precise Multimodal Prediction

Implicit Latent Variable Model for Scene-Consistent Motion Forecasting

End-to-end Contextual Perception and Prediction with Interaction Transformer

Testing the Safety of Self-driving Vehicles by Simulating Perception and Prediction

Learning Lane Graph Representations for Motion Forecasting

RadarNet: Exploiting Radar for Robust Perception of Dynamic Objects

Perceive, Predict, and Plan: Safe Motion Planning Through Interpretable Semantic Representations

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

Physically Realizable Adversarial Examples for LiDAR Detection

PolyTransform: Deep Polygon Transformer for Instance Segmentation

PnPNet: End-to-End Perception and Prediction with Tracking in the Loop

Discrete Residual Flow for Probabilistic Pedestrian Behavior Prediction

Identifying Unknown Instances for Autonomous Driving

DeepPruner: Learning Efficient Stereo Matching via Differentiable PatchMatch

Learning Joint 2D-3D Representations for Depth Completion

End-to-End Interpretable Neural Motion Planner

UPSNet: A Unified Panoptic Segmentation Network

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

Deep Rigid Instance Scene Flow

DeepSignals: Predicting Intent of Drivers Through Visual Signals

Deep Multi-Sensor Lane Detection

HDNET: Exploiting HD Maps for 3D Object Detection

IntentNet: Learning to Predict Intention from Raw Sensor Data

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

Deep Continuous Fusion for Multi-Sensor 3D Object Detection

SBNet: Sparse Blocks Network for Fast Inference

Deep Parametric Continuous Convolutional Neural Networks

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

Matching Adversarial Networks

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

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

R&D Homepage