SynDiff-AD: Improving Semantic Segmentation and End-to-End Autonomous Driving with Synthetic Data from Latent Diffusion Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Goel, Harsh, Narasimhan, Sai Shankar, Akcin, Oguzhan, Chinchali, Sandeep
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914094355841024
author Goel, Harsh
Narasimhan, Sai Shankar
Akcin, Oguzhan
Chinchali, Sandeep
author_facet Goel, Harsh
Narasimhan, Sai Shankar
Akcin, Oguzhan
Chinchali, Sandeep
contents In recent years, significant progress has been made in collecting large-scale datasets to improve segmentation and autonomous driving models. These large-scale datasets are often dominated by common environmental conditions such as "Clear and Day" weather, leading to decreased performance in under-represented conditions like "Rainy and Night". To address this issue, we introduce SynDiff-AD, a novel data augmentation pipeline that leverages diffusion models (DMs) to generate realistic images for such subgroups. SynDiff-AD uses ControlNet-a DM that guides data generation conditioned on semantic maps-along with a novel prompting scheme that generates subgroup-specific, semantically dense prompts. By augmenting datasets with SynDiff-AD, we improve the performance of segmentation models like Mask2Former and SegFormer by up to 1.2% and 2.3% on the Waymo dataset, and up to 1.4% and 0.7% on the DeepDrive dataset, respectively. Additionally, we demonstrate that our SynDiff-AD pipeline enhances the driving performance of end-to-end autonomous driving models, like AIM-2D and AIM-BEV, by up to 20% across diverse environmental conditions in the CARLA autonomous driving simulator, providing a more robust model. We release our code and pipeline at https://github.com/UTAustin-SwarmLab/SynDiff-AD.
format Preprint
id arxiv_https___arxiv_org_abs_2411_16776
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SynDiff-AD: Improving Semantic Segmentation and End-to-End Autonomous Driving with Synthetic Data from Latent Diffusion Models
Goel, Harsh
Narasimhan, Sai Shankar
Akcin, Oguzhan
Chinchali, Sandeep
Computer Vision and Pattern Recognition
In recent years, significant progress has been made in collecting large-scale datasets to improve segmentation and autonomous driving models. These large-scale datasets are often dominated by common environmental conditions such as "Clear and Day" weather, leading to decreased performance in under-represented conditions like "Rainy and Night". To address this issue, we introduce SynDiff-AD, a novel data augmentation pipeline that leverages diffusion models (DMs) to generate realistic images for such subgroups. SynDiff-AD uses ControlNet-a DM that guides data generation conditioned on semantic maps-along with a novel prompting scheme that generates subgroup-specific, semantically dense prompts. By augmenting datasets with SynDiff-AD, we improve the performance of segmentation models like Mask2Former and SegFormer by up to 1.2% and 2.3% on the Waymo dataset, and up to 1.4% and 0.7% on the DeepDrive dataset, respectively. Additionally, we demonstrate that our SynDiff-AD pipeline enhances the driving performance of end-to-end autonomous driving models, like AIM-2D and AIM-BEV, by up to 20% across diverse environmental conditions in the CARLA autonomous driving simulator, providing a more robust model. We release our code and pipeline at https://github.com/UTAustin-SwarmLab/SynDiff-AD.
title SynDiff-AD: Improving Semantic Segmentation and End-to-End Autonomous Driving with Synthetic Data from Latent Diffusion Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2411.16776