AI Paper Atlas

Research map / Point clouds / Point cloud classification

DAP-MAE: Domain-Adaptive Point Cloud Masked Autoencoder for Effective Cross-Domain Learning

Ziqi Gao, Qiufu Li, Linlin Shen

ICCV 2025 · Point clouds › Point cloud classification · approach: Masked point autoencoders · 3 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 cross-domain pretraining data by simple combination (ablation)ReCon-SMC · ScanObjectNN OBJ_BGaccuracy: 94.15 -> 94.32No clear effect
add cross-domain data by simple combinationReCon-SMC · BU3DFE expression recognitionaccuracy: 89.13 -> 88.52Hurt
add cross-domain data by simple combinationReCon-SMC · Bosphorus expression recognitionaccuracy: 87.69 -> 87.23Hurt
add cross-domain data by simple combinationReCon-SMC · ScanNetV2 detectionAP50: 42.7 -> 42.5Hurt
add heterogeneous domain adapter (ablation)DAP-MAE with cross-domain data · ScanObjectNN OBJ_BGaccuracy: 94.32 -> 94.66Improved
add domain feature generator (ablation)DAP-MAE with cross-domain data · ScanObjectNN OBJ_BGaccuracy: 94.32 -> 94.66Improved
add both HDA and DFG (ablation)DAP-MAE with cross-domain data · ScanObjectNN OBJ_BGaccuracy: 94.32 -> 95.18Improved
add fusion of other-domain MLPs in HDA (ablation)DAP-MAE · ScanObjectNN OBJ_BGaccuracy: 94.66 -> 95.18Improved
use MLP instead of plain adding to fuse domains O+F (ablation)DAP-MAE · ScanObjectNN OBJ_BGaccuracy: 92.59 -> 93.80Improved
use MLP instead of FC to predict fusion coefficients for O+F+S (ablation)DAP-MAE · ScanObjectNN OBJ_BGaccuracy: 93.80 -> 95.18Improved
combine class, domain and point cloud features in head (ablation)DAP-MAE · ScanObjectNN OBJ_BGaccuracy: 94.49 -> 95.18Improved
use loss weights 100/0.001 instead of 1.0/1.0 (ablation)DAP-MAE · ScanObjectNN OBJ_BGaccuracy: 93.80 -> 95.18Improved

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