Saved in:
Bibliographic Details
Main Authors: Yang, Jiezhi, Desai, Khushi, Packer, Charles, Bhatia, Harshil, Rhinehart, Nicholas, McAllister, Rowan, Gonzalez, Joseph
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2401.18075
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929428197539840
author Yang, Jiezhi
Desai, Khushi
Packer, Charles
Bhatia, Harshil
Rhinehart, Nicholas
McAllister, Rowan
Gonzalez, Joseph
author_facet Yang, Jiezhi
Desai, Khushi
Packer, Charles
Bhatia, Harshil
Rhinehart, Nicholas
McAllister, Rowan
Gonzalez, Joseph
contents We propose CARFF, a method for predicting future 3D scenes given past observations. Our method maps 2D ego-centric images to a distribution over plausible 3D latent scene configurations and predicts the evolution of hypothesized scenes through time. Our latents condition a global Neural Radiance Field (NeRF) to represent a 3D scene model, enabling explainable predictions and straightforward downstream planning. This approach models the world as a POMDP and considers complex scenarios of uncertainty in environmental states and dynamics. Specifically, we employ a two-stage training of Pose-Conditional-VAE and NeRF to learn 3D representations, and auto-regressively predict latent scene representations utilizing a mixture density network. We demonstrate the utility of our method in scenarios using the CARLA driving simulator, where CARFF enables efficient trajectory and contingency planning in complex multi-agent autonomous driving scenarios involving occlusions.
format Preprint
id arxiv_https___arxiv_org_abs_2401_18075
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CARFF: Conditional Auto-encoded Radiance Field for 3D Scene Forecasting
Yang, Jiezhi
Desai, Khushi
Packer, Charles
Bhatia, Harshil
Rhinehart, Nicholas
McAllister, Rowan
Gonzalez, Joseph
Computer Vision and Pattern Recognition
We propose CARFF, a method for predicting future 3D scenes given past observations. Our method maps 2D ego-centric images to a distribution over plausible 3D latent scene configurations and predicts the evolution of hypothesized scenes through time. Our latents condition a global Neural Radiance Field (NeRF) to represent a 3D scene model, enabling explainable predictions and straightforward downstream planning. This approach models the world as a POMDP and considers complex scenarios of uncertainty in environmental states and dynamics. Specifically, we employ a two-stage training of Pose-Conditional-VAE and NeRF to learn 3D representations, and auto-regressively predict latent scene representations utilizing a mixture density network. We demonstrate the utility of our method in scenarios using the CARLA driving simulator, where CARFF enables efficient trajectory and contingency planning in complex multi-agent autonomous driving scenarios involving occlusions.
title CARFF: Conditional Auto-encoded Radiance Field for 3D Scene Forecasting
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2401.18075