AI Paper Atlas

Research map / 3D vision / 3D generation

MvDeCor: Multi-View Dense Correspondence Learning for Fine-Grained 3D Segmentation

Gopal Sharma, Kangxue Yin, Subhransu Maji, Evangelos Kalogerakis, Or Litany, Sanja Fidler

ECCV 2022 · 3D vision › 3D generation · approach: Learning 3D reconstruction from images · 10 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 multi-view dense correspondence pretraining vs training from scratch2D CNN · PartNet (k=10)part mIoU: 32.0 -> 35.9Improved
add multi-view dense correspondence pretraining vs DenseCL2D CNN · PartNet (k=10)part mIoU: 34.2 -> 35.9Improved
use ImageNet pretrained weights vs scratch2D CNN · PartNet (k=10)part mIoU: 32.0 -> 32.8Improved
use 2D dense contrastive learning (DenseCL) vs ImageNet init2D CNN · PartNet (k=10)part mIoU: 32.8 -> 34.2Improved
3D scratch vs 2D scratchsegmentation network · PartNet (k=10)part mIoU: 32.0 -> 30.3Hurt
add learned deformation-based alignment (3D)3D network · PartNet (k=10)part mIoU: 30.3 -> 27.5Hurt
add MvDeCor pretraining with sparse labeled views2D CNN · PartNet (k=10, v=5)part mIoU: 25.9 -> 30.3Improved
add MvDeCor pretraining on larger label set2D CNN · PartNet chair/table/lamp (k=30)part mIoU: 13.7 -> 16.6Improved
add MvDeCor pretraining on RenderPeople RGB2D CNN · RenderPeople RGB (k=5)part mIoU: 50.2 -> 67.5Improved
add MvDeCor pretraining with geometry input2D CNN · RenderPeople RGB+Geom (k=5)part mIoU: 55.3 -> 58.8Improved
add MvDeCor pretraining with 3 labeled views and geometry2D CNN · RenderPeople RGB+Geom (k=5, v=3)part mIoU: 40.6 -> 50.3Improved
use ImageNet init with depth and normal input2D CNN · RenderPeople RGB+Geom (k=5)part mIoU: 55.3 -> 55.3No clear effect

Later papers that compared with this method

1 later accepted paper reports MvDeCor as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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