CARLANE: A Lane Detection Benchmark for Unsupervised Domain Adaptation from Simulation to multiple Real-World Domains

Julian Gebele*,1, Bonifaz Stuhr*,1,2, Johann Haselberger*,1,3
*Equal Contribution, 1University of Applied Science Kempten, 2Autonomous University of Barcelona, 3Technische Universität Berlin
Dataset Tutorial Paper Video (on SlidesLive) Dataset GitHub Baselines GitHub Papers With Code

MoLane

Adaption from Simulation to Model Vehicle


TuLane

Adaption from Simulation to Real Scenarios


MuLane

Multi-Target Adaption from Simulation to Both


Abstract

Unsupervised Domain Adaptation demonstrates great potential to mitigate domain shifts by transferring models from labeled source domains to unlabeled target domains. While Unsupervised Domain Adaptation has been applied to a wide variety of complex vision tasks, only few works focus on lane detection for autonomous driving. This can be attributed to the lack of publicly available datasets. To facilitate research in these directions, we propose CARLANE, a 3-way sim-to-real domain adaptation benchmark for 2D lane detection. CARLANE encompasses the single-target datasets MoLane and TuLane and the multi-target dataset MuLane. These datasets are built from three different domains, which cover diverse scenes and contain a total of 163K unique images, 118K of which are annotated. In addition we evaluate and report systematic baselines, including our own method, which builds upon Prototypical Cross-domain Self-supervised Learning. We find that false positive and false negative rates of the evaluated domain adaptation methods are high compared to those of fully supervised baselines. This affirms the need for benchmarks such as CARLANE to further strengthen research in Unsupervised Domain Adaptation for lane detection. CARLANE, all evaluated models and the corresponding implementations are publicly available at https://carlanebenchmark.github.io.

Lane Distribution

Dataset Details

Dataset Domains Total Images Train Validation Test Lanes
MoLane

CARLA Simulation

Model Vehicle

84,000

46,843

80,000

43,843*

4,000

2,000

-

1,000

≤ 2

≤ 2
TuLane

CARLA Simulation

TuSimple

26,400

6,408

24,000

3,268

2,400

358

-

2,782

≤ 4

≤ 4
MuLane

CARLA Simulation

Model Vehicle+TuSimple

52,800

12,536

48,000

6,536**

4,800

4,000

-

2,000

≤ 4

≤ 4

*unlabeled **partially labeled

Baselines

MoLane


TuLane


MuLane

t-SNE

Weights

Model MoLane TuLane MuLane
UFLD-SO UFLD-SO_ResNet18
UFLD-SO_ResNet34 		UFLD-SO_ResNet18
UFLD-SO_ResNet34 		UFLD-SO_ResNet18
UFLD-SO_ResNet34
DANN DANN_ResNet18
DANN_ResNet34 		DANN_ResNet18
DANN_ResNet34 		DANN_ResNet18
DANN_ResNet34
ADDA ADDA_ResNet18
ADDA_ResNet34 		ADDA_ResNet18
ADDA_ResNet34 		ADDA_ResNet18
ADDA_ResNet34
SGADA SGADA_ResNet18
SGADA_ResNet34 		SGADA_ResNet18
SGADA_ResNet34 		SGADA_ResNet18
SGADA_ResNet34
SGPCS SGPCS_ResNet18
SGPCS_ResNet34 		SGPCS_ResNet18
SGPCS_ResNet34 		SGPCS_ResNet18
SGPCS_ResNet34
UFLD-TO UFLD-TO_ResNet18
UFLD-TO_ResNet34 		UFLD-TO_ResNet18
UFLD-TO_ResNet34 		UFLD-TO_ResNet18
UFLD-TO_ResNet34

BibTeX

@inproceedings{NEURIPS2022_19a26064,
 author = {Stuhr, Bonifaz and Haselberger, Johann and Gebele, Julian},
 booktitle = {Advances in Neural Information Processing Systems},
 editor = {S. Koyejo and S. Mohamed and A. Agarwal and D. Belgrave and K. Cho and A. Oh},
 pages = {4046--4058},
 publisher = {Curran Associates, Inc.},
 title = {CARLANE: A Lane Detection Benchmark for Unsupervised Domain Adaptation from Simulation to multiple Real-World Domains},
 volume = {35},
 year = {2022}
}

Dataset DOI

10.34740/kaggle/dsv/3798459

License


CARLANE is licensed under the
 Apache License 2.0

A permissive license whose main conditions require preservation of copyright and license notices. Contributors provide an express grant of patent rights. Licensed works, modifications, and larger works may be distributed under different terms and without source code.


Permissions
Commercial use
 Modification
 Distribution
 Patent use
 Private use

Limitations
Trademark use
 Liability
 Warranty

Conditions
License and copyright notice  
 State changes 

 Apache License Version 2.0, January 2004
            
            Copyright 2022 CARLANE Team (Julian Gebele, Bonifaz Stuhr, Johann Haselberger)

            Licensed under the Apache License, Version 2.0 (the "License");
            you may not use this file except in compliance with the License.
            You may obtain a copy of the License at
         
                http://www.apache.org/licenses/LICENSE-2.0
         
            Unless required by applicable law or agreed to in writing, software
            distributed under the License is distributed on an "AS IS" BASIS,
            WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
            See the License for the specific language governing permissions and
            limitations under the License.