AI Paper Atlas

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

360SFUDA++: Towards Source-Free UDA for Panoramic Segmentation by Learning Reliable Category Prototypes

Zheng Xu, Pengyuan Zhou, Athanasios V. Vasilakos, Lin Wang

TPAMI 2024 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Adaptation for semantic segmentation · 18 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 reliable prototype adaptation (L_un + L_ppa) to supervised source loss (ablation)360SFUDA++ · Cityscapes-to-DensePASSmIoU: 38.65 -> 51.05Improved
add L_un to supervised loss (ablation)360SFUDA++ · Cityscapes-to-DensePASSmIoU: 38.65 -> 43.05Improved
add L_ppa on top of L_sup and L_un (ablation)360SFUDA++ · Cityscapes-to-DensePASSmIoU: 43.05 -> 51.05Improved
add source-model fine-tuning loss L_sft (ablation)360SFUDA++ · Cityscapes-to-DensePASSmIoU: 51.05 -> 51.16Improved
add cross-dual-attention module (CDAM) alone (ablation)360SFUDA++ · Cityscapes-to-DensePASSmIoU: 38.65 -> 44.24Improved
add batch-norm statistics loss L_bns to CDAM (ablation)360SFUDA++ · Cityscapes-to-DensePASSmIoU: 44.24 -> 44.79Improved
use RP2AM instead of PPAM (ablation)360SFUDA++ · Cityscapes-to-DensePASSmIoU: 45.42 -> 51.05Improved
use cross-projection dual attention instead of within-feature dual attention (ablation)CDAM · Cityscapes-to-DensePASSmIoU: 41.53 -> 44.24Improved
use fixed FoV projection of 90 degrees versus none (ablation)360SFUDA++ · Cityscapes-to-DensePASSmIoU: 38.65 -> 44.28Improved
enlarge FFP field of view from 90 to 360 degrees (ablation)360SFUDA++ · Cityscapes-to-DensePASSmIoU: 44.28 -> 40.31Hurt
raise CDAM trade-off gamma from 0.1 to 0.2 (ablation)CDAM · Cityscapes-to-DensePASSmIoU: 44.24 -> 43.07Hurt
upgrade 360SFUDA to 360SFUDA++ with SegFormer-B2360SFUDA · Cityscapes-to-DensePASSmIoU: 50.12 -> 52.99Improved

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