AI Paper Atlas

Research map / Point clouds / Point cloud classification

Multi-modal Relation Distillation for Unified 3D Representation Learning

Huiqun Wang, Yiping Bao, Panwang Pan, Zeming Li, Xiao Liu, Ruijie Yang, Di Huang

ECCV 2024 · Point clouds › Point cloud classification · approach: Point-language and open-vocabulary 3D · 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 relation distillation with normalized similarity (intra-modal) (ablation)contrastive 3D-image-text alignment · ModelNet40 zero-shotTop-1 accuracy (%): 72.2 -> 73.1Improved
use partial-order relation representation (intra-modal) instead of normalized similarity (ablation)relation distillation · Objaverse-LVIS zero-shotTop-1 accuracy (%): 11.0 -> 10.0Hurt
add both intra- and cross-modal distillation without dynamic weights (ablation)MRD · ScanObjectNN zero-shotTop-1 accuracy (%): 54.6 -> 53.5 (vs intra only)Mixed
add dynamic distillation weights (learnable softmax weights) (ablation)MRD with intra and cross relations · ModelNet40 zero-shotTop-1 accuracy (%): 73.0 -> 74.2Improved
set loss weight lambda to 3 versus 5 (ablation)MRD · ModelNet40 zero-shotTop-1 accuracy (%): 72.4 -> 74.2Improved
scale PointBERT backbone from 5M to 307M with MRD (ablation)MRD · Objaverse-LVIS zero-shotTop-1 accuracy (%): 47.6 -> 53.6Improved
scale MRD-B to MRD-L on ModelNet40 (ablation)MRD · ModelNet40 zero-shotTop-1 accuracy (%): 88.8 -> 87.3Hurt
pretrain with MRD instead of ULIP before fine-tuningPointBERT-22M · ScanObjectNN fine-tuningInstance accuracy (%): 86.4 -> 89.6Improved
use MRD instead of OpenShape/Uni3D for text-to-3D retrievalCap3D to Objaverse retrieval · instance-wise retrievalTop-1 accuracy (%): 18.9 -> 20.5Improved
use ULIP2 image-text data instead of default data for MRD training (ablation)MRD-T and MRD-S on ShapeNet · Objaverse zero-shotaccuracyImproved

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