AI Paper Atlas

Research map / Detection and segmentation / CNN features

Ultra Fast Deep Lane Detection With Hybrid Anchor Driven Ordinal Classification

Zequn Qin, Pengyi Zhang, Xi Li

TPAMI 2022 · Detection and segmentation › CNN features · approach: Lane detection · 230 citations (OpenAlex)

Read the paperCheck 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
use hybrid anchors instead of row anchors only (ablation)UFLDv2 (ResNet-18) · CULaneF1 (%): 66.09 -> 74.70Improved
use hybrid anchors instead of column anchors only (ablation)UFLDv2 (ResNet-18) · CULaneF1 (%): 57.29 -> 74.70Improved
use ordinal classification instead of regression (ablation)UFLDv2 (ResNet-18) · CULaneF1 (%): 64.29 -> 74.70Improved
use ordinal classification (expectation, three losses) instead of standard classification (ablation)UFLDv2 (ResNet-18) · CULaneF1 (%): 67.24 -> 74.70Improved
add expectation loss to base cross-entropy loss (ablation)UFLDv2 (ResNet-18) · CULaneF1 (%): 74.18 -> 74.70Improved
use expectation loss alone instead of base loss (ablation)UFLDv2 (ResNet-18) · CULaneF1 (%): 74.18 -> 69.70Hurt
add feature level test time augmentation (ablation)UFLDv2 (ResNet-18) · CULaneF1 total (%): 74.7 -> 75.0Improved
add feature level test time augmentation (ablation)UFLDv2 (ResNet-34) · CULaneF1 total (%): 75.9 -> 76.0No clear effect
use UFLDv2 (hybrid anchor, ordinal classification) over original UFLDResNet-18 · CULaneF1 total (%): 68.4 -> 74.7Improved
use classification formulation instead of segmentation (ResNet-18)UFLDv2 · TuSimpleaccuracy (%): 92.69 -> 95.84Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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