AI Paper Atlas

Research map / Detection and segmentation / CNN features

FastDraw: Addressing the Long Tail of Lane Detection by Adapting a Sequential Prediction Network

Jonah Philion

CVPR 2019 · Detection and segmentation › CNN features · approach: Lane detection · 203 citations (OpenAlex)

Check your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train with MUNIT style-transferred images (ablation)FastDraw ResNet-50 · TuSimple validationaccuracy (%): 94.9 -> 95.2Improved
train with MUNIT style-transferred images (ablation)FastDraw ResNet-50 · TuSimple validationFP: 0.059 -> 0.076Hurt
train with MUNIT style-transferred images (ablation)FastDraw ResNet-50 · TuSimple validationFN: 0.052 -> 0.045Improved
use ResNet-18 rather than ResNet-50 as the backbone (ablation)FastDraw · TuSimple validationaccuracy (%): 94.9 -> 94.9No clear effect
use FastDraw instead of the ResNet-50 baselineCULane ResNet-50 baseline · CULane test, NormalIoU at overlap 0.5: 87.4 -> 85.9Hurt
use FastDraw instead of the ResNet-50 baselineCULane ResNet-50 baseline · CULane test, CrowdedIoU at overlap 0.5: 64.1 -> 63.6Hurt
use FastDraw instead of the ResNet-50 baselineCULane ResNet-50 baseline · CULane test, NightIoU at overlap 0.5: 60.6 -> 57.8Hurt
use FastDraw instead of the ResNet-50 baselineCULane ResNet-50 baseline · CULane test, No lineIoU at overlap 0.5: 38.1 -> 40.6Improved
use FastDraw instead of the ResNet-50 baselineCULane ResNet-50 baseline · CULane test, ShadowIoU at overlap 0.5: 60.7 -> 59.9Hurt
use FastDraw instead of the ResNet-50 baselineCULane ResNet-50 baseline · CULane test, ArrowIoU at overlap 0.5: 79.0 -> 79.4Improved
use FastDraw instead of the ResNet-50 baselineCULane ResNet-50 baseline · CULane test, DazzleIoU at overlap 0.5: 54.1 -> 57.0Improved
use FastDraw instead of the ResNet-50 baselineCULane ResNet-50 baseline · CULane test, CurveIoU at overlap 0.5: 59.8 -> 65.2Improved

Later papers that compared with this method

1 later accepted paper reports FastDraw as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy