AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

2PCNet: Two-Phase Consistency Training for Day-to-Night Unsupervised Domain Adaptive Object Detection

Mikhail Kennerley, Jian-Gang Wang, Bharadwaj Veeravalli, Robby T. Tan

CVPR 2023 · Detection and segmentation › Open-set and few-shot detection · approach: Domain-adaptive object detection · 76 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 Two-Phase Consistency, matching student proposals with teacher predictions and combining high- and lower-confidence pseudo-labels (ablation)Mean Teacher on BDD100K · BDD100K nightAP: 31.7 -> 45.2Improved
Add NightAug to Two-Phase Consistency (ablation)2PCNet on BDD100K · BDD100K nightAP_s: 8.2 -> 8.6Improved
Add NightAug to Two-Phase Consistency (ablation)2PCNet on BDD100K · BDD100K nightAP: 45.2 -> 45.8Improved
Add student-scaling to Two-Phase Consistency and NightAug (ablation)2PCNet on BDD100K · BDD100K nightAP_s: 8.6 -> 9.1Improved
Add student-scaling to Two-Phase Consistency and NightAug (ablation)2PCNet on BDD100K · BDD100K nightAP: 45.8 -> 46.4Improved
Add a domain classifier to Two-Phase Consistency, NightAug, and student-scaling (ablation)2PCNet on BDD100K · BDD100K nightAP: 46.4 -> 44.5Hurt
Add student-scaling to Two-Phase Consistency and NightAug (ablation)2PCNet on BDD100K · BDD100K nightAP_l: 42.2 -> 41.7Hurt
Use 2PCNet rather than the strongest listed domain-adaptation baselineDay-to-night detection · BDD100K nightAP: 41.3 -> 46.4Improved
Use 2PCNet rather than the strongest listed domain-adaptation baselineDay-to-night detection · SHIFT night, clearAP: 43.7 -> 49.1Improved
Use 2PCNet rather than the strongest listed image-translation baselineDay-to-night detection · BDD100K nightAP_coco: 22.9 -> 23.5Improved

Later papers that compared with this method

2 later accepted papers report 2PCNet as a baseline in their results tables; 2 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