SLPC: a VRNN-based approach for stochastic lidar prediction and completion in autonomous driving

02/19/2021
by   George Eskandar, et al.
0

Predicting future 3D LiDAR pointclouds is a challenging task that is useful in many applications in autonomous driving such as trajectory prediction, pose forecasting and decision making. In this work, we propose a new LiDAR prediction framework that is based on generative models namely Variational Recurrent Neural Networks (VRNNs), titled Stochastic LiDAR Prediction and Completion (SLPC). Our algorithm is able to address the limitations of previous video prediction frameworks when dealing with sparse data by spatially inpainting the depth maps in the upcoming frames. Our contributions can thus be summarized as follows: we introduce the new task of predicting and completing depth maps from spatially sparse data, we present a sparse version of VRNNs and an effective self-supervised training method that does not require any labels. Experimental results illustrate the effectiveness of our framework in comparison to the state of the art methods in video prediction.

READ FULL TEXT

page 2

page 4

research
12/08/2021

FPPN: Future Pseudo-LiDAR Frame Prediction for Autonomous Driving

LiDAR sensors are widely used in autonomous driving due to the reliable ...
research
07/01/2018

Self-supervised Sparse-to-Dense: Self-supervised Depth Completion from LiDAR and Monocular Camera

Depth completion, the technique of estimating a dense depth image from s...
research
02/14/2019

Sparse and noisy LiDAR completion with RGB guidance and uncertainty

This work proposes a new method to accurately complete sparse LiDAR maps...
research
04/06/2020

CVPR 2019 WAD Challenge on Trajectory Prediction and 3D Perception

This paper reviews the CVPR 2019 challenge on Autonomous Driving. Baidu'...
research
02/08/2022

GLPU: A Geometric Approach For Lidar Pointcloud Upsampling

In autonomous driving, lidar is inherent for the understanding of the 3D...
research
12/17/2020

Temporal LiDAR Frame Prediction for Autonomous Driving

Anticipating the future in a dynamic scene is critical for many fields s...
research
11/04/2019

Disentangling Human Dynamics for Pedestrian Locomotion Forecasting with Noisy Supervision

We tackle the problem of Human Locomotion Forecasting, a task for jointl...

Please sign up or login with your details

Forgot password? Click here to reset