AI Paper Atlas

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

MICDrop: Masking Image and Depth Features via Complementary Dropout for Domain-Adaptive Semantic Segmentation

Linyan Yang, Lukas Hoyer, Mark Weber, Tobias Fischer, Dengxin Dai, Laura Leal-Taixé, Daniel Cremers, Marc Pollefeys, Luc Van Gool

ECCV 2024 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Adaptation for semantic segmentation · 17 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 depth input with fusion module (no masking) (ablation)DAFormer · GTA->CityscapesmIoU: 68.3 -> 69.1Improved
add complementary masking of RGB and depth features across all levels (ablation)DAFormer + depth · GTA->CityscapesmIoU: 69.1 -> 70.1Improved
use independent masking of RGB and depth instead of complementary (ablation)DAFormer + depth · GTA->CityscapesmIoU: 69.1 -> 69.1No clear effect
sample complementary mask independently at each feature level (ablation)MICDrop · GTA->CityscapesmIoU: 70.1 -> 69.7Hurt
use combined local self-attention and global cross-attention fusion (ablation)MICDrop fusion module · GTA->CityscapesmIoU: 69.7 -> 70.1 (vs local only)Improved
use global depth-guided cross-attention alone for fusion (ablation)MICDrop fusion module · GTA->CityscapesmIoU: 68.3 -> 68.1 (vs no-depth baseline)Hurt
use CMX cross-attention fusion instead of proposed fusion (ablation)MICDrop · GTA->CityscapesmIoU: 70.1 -> 68.6Hurt
freeze the pretrained RGB encoder (ablation)DAFormer + depth · GTA->CityscapesmIoU: 68.7 -> 69.1Improved
add a full unmasked forward pass alongside masked mix-batch (ablation)MICDrop training · GTA->CityscapesmIoU: 67.8 -> 70.1Improved
also mask source-batch features in the loss (ablation)MICDrop training · GTA->CityscapesmIoU: 70.1 -> 69.9Hurt
increase mask patch size from 16 to 64 (ablation)MICDrop · GTA->CityscapesmIoU: 69.4 -> 70.1Improved
plug MICDrop into MIC_HRDAMIC_HRDA · GTA->CityscapesmIoU: 75.9 -> 76.6Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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