Papers › CARLANE: A Lane Detection Benchmark for Unsupervised Domain Adaptation from Simulation...
CARLANE: A Lane Detection Benchmark for Unsupervised Domain Adaptation from Simulation to multiple Real-World Domains
Julian Gebele, Bonifaz Stuhr, Johann Haselberger
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.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Datasets
Introduced by this paper, per the archive.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Domain Adaptation | MoLane | UFLD-SGPCS-ResNet18 | Lane Accuracy (LA) | 93.94 | #1 of 8 | Archive leaderboard | report |
| Domain Adaptation | MoLane | UFLD-SGADA-ResNet18 | Lane Accuracy (LA) | 93.82 | #2 of 8 | Archive leaderboard | report |
| Domain Adaptation | MoLane | UFLD-SGPCS-ResNet32 | Lane Accuracy (LA) | 93.53 | #3 of 8 | Archive leaderboard | report |
| Domain Adaptation | MoLane | UFLD-SGADA-ResNet32 | Lane Accuracy (LA) | 93.31 | #4 of 8 | Archive leaderboard | report |
| Domain Adaptation | MoLane | UFLD-ADDA-ResNet18 | Lane Accuracy (LA) | 92.85 | #5 of 8 | Archive leaderboard | report |
| Domain Adaptation | MoLane | UFLD-ADDA-ResNet32 | Lane Accuracy (LA) | 92.39 | #6 of 8 | Archive leaderboard | report |
| Domain Adaptation | MoLane | UFLD-DANN-ResNet32 | Lane Accuracy (LA) | 90.91 | #7 of 8 | Archive leaderboard | report |
| Domain Adaptation | MoLane | UFLD-DANN-ResNet18 | Lane Accuracy (LA) | 87.65 | #8 of 8 | Archive leaderboard | report |
| Domain Adaptation | MuLane | UFLD-SGADA-ResNet32 | Lane Accuracy (LA) | 91.63 | #1 of 8 | Archive leaderboard | report |
| Domain Adaptation | MuLane | UFLD-SGPCS-ResNet18 | Lane Accuracy (LA) | 91.57 | #2 of 8 | Archive leaderboard | report |
| Domain Adaptation | MuLane | UFLD-SGPCS-ResNet32 | Lane Accuracy (LA) | 91.55 | #3 of 8 | Archive leaderboard | report |
| Domain Adaptation | MuLane | UFLD-SGADA-ResNet18 | Lane Accuracy (LA) | 90.71 | #4 of 8 | Archive leaderboard | report |
| Domain Adaptation | MuLane | UFLD-ADDA-ResNet32 | Lane Accuracy (LA) | 90.22 | #5 of 8 | Archive leaderboard | report |
| Domain Adaptation | MuLane | UFLD-ADDA-ResNet18 | Lane Accuracy (LA) | 89.83 | #6 of 8 | Archive leaderboard | report |
| Domain Adaptation | MuLane | UFLD-DANN-ResNet32 | Lane Accuracy (LA) | 88.76 | #7 of 8 | Archive leaderboard | report |
| Domain Adaptation | MuLane | UFLD-DANN-ResNet18 | Lane Accuracy (LA) | 86.01 | #8 of 8 | Archive leaderboard | report |
| Domain Adaptation | TuLane | UFLD-SGPCS-ResNet32 | Lane Accuracy (LA) | 93.29 | #1 of 8 | Archive leaderboard | report |
| Domain Adaptation | TuLane | UFLD-SGADA-ResNet32 | Lane Accuracy (LA) | 92.04 | #2 of 8 | Archive leaderboard | report |
| Domain Adaptation | TuLane | UFLD-SGADA-ResNet18 | Lane Accuracy (LA) | 91.70 | #3 of 8 | Archive leaderboard | report |
| Domain Adaptation | TuLane | UFLD-SGPCS-ResNet18 | Lane Accuracy (LA) | 91.55 | #4 of 8 | Archive leaderboard | report |
| Domain Adaptation | TuLane | UFLD-ADDA-ResNet32 | Lane Accuracy (LA) | 91.39 | #5 of 8 | Archive leaderboard | report |
| Domain Adaptation | TuLane | UFLD-DANN-ResNet32 | Lane Accuracy (LA) | 91.06 | #6 of 8 | Archive leaderboard | report |
| Domain Adaptation | TuLane | UFLD-ADDA-ResNet18 | Lane Accuracy (LA) | 90.72 | #7 of 8 | Archive leaderboard | report |
| Domain Adaptation | TuLane | UFLD-DANN-ResNet18 | Lane Accuracy (LA) | 88.74 | #8 of 8 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Introduced by this paper: SGPCS
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections