AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Train Till You Drop: Towards Stable and Robust Source-free Unsupervised 3D Domain Adaptation

Björn Michele, Alexandre Boulch, Tuan-Hung VU, Gilles Puy, Renaud Marlet, Nicolas Courty

ECCV 2024 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Adaptation for semantic segmentation · 4 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
add TTYD_core (entropy + source-class-distribution loss, BN-to-linear adaptation, agreement-based stopping) vs best hyperparameter-free baselineMinkowski U-Net SFUDA · nuScenes->SemanticKITTImIoU: 41.7 -> 44.5Improved
add self-training module on top of TTYD_coreTTYD_core · nuScenes->SemanticKITTImIoU: 44.5 -> 45.4Improved
use agreement-with-reference stopping criterion instead of entropy validator (ablation)TTYD training scheme · nuScenes->SemanticPOSSmIoU: 23.7 -> 61.1Improved
add stopping criterion to training scheme that otherwise degrades (ablation)TTYD training scheme · nuScenes->SemanticPOSSmIoU: -38.0 pp drop without stoppingImproved
add source-class-distribution loss (ablation)TTYD training scheme · nuScenes->SemanticKITTImax mIoU: 34.4 -> 40.9Improved
add entropy discriminability loss alongside distribution loss (ablation)TTYD training scheme · nuScenes->SemanticKITTImax mIoU: 40.9 -> 44.7Improved
add entropy loss alone, or uniform-distribution loss alone (ablation)TTYD training scheme · nuScenes->SemanticKITTImax mIoU: 34.4 -> 34.4No clear effect
use source class distribution instead of uniform as prior (ablation)TTYD training scheme · nuScenes->SemanticKITTImax mIoU: 35.0 -> 44.7Improved
adapt only the classification layer (ablation)TTYD training scheme · nuScenes->SemanticKITTImax mIoU: 44.7 -> 34.4Hurt
adapt BN running statistics online at train and eval time (ablation)TTYD training scheme · nuScenes->SemanticKITTImax mIoU: 44.7 -> 39.1Hurt
use MeanBN rather than PTBN as the reference model for stopping (ablation)TTYD_stop · nuScenes->SemanticKITTImIoU: 44.5 -> 39.0Hurt
use L1, L2 or symmetric KL instead of hard agreement counting (ablation)TTYD_stop · nuScenes->SemanticKITTImIoU: 44.5 -> 44.5No clear effect

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