AI Paper Atlas

Research map / Point clouds / Point cloud classification

WildRefer: 3D Object Localization in Large-scale Dynamic Scenes with Multi-modal Visual Data and Natural Language

Zhenxiang Lin, Xidong Peng, Peishan Cong, Ge Zheng, Yujing Sun, Yuenan HOU, Xinge Zhu, Sibei Yang, Yuexin Ma

ECCV 2024 · Point clouds › Point cloud classification · approach: Point-language and open-vocabulary 3D · 14 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 Dynamic Visual Encoder (temporal attention) to the baseline (ablation)WildRefer baseline · STReferAcc@0.5: 45.12 -> 50.28Improved
add Triple-modal Feature Interaction (image fusion) on top of DVE (ablation)WildRefer · STReferAcc@0.5: 50.28 -> 54.97Improved
add DVE to baseline (ablation)WildRefer baseline · LifeReferAcc@0.25: 32.46 -> 38.27Improved
add TFI on top of DVE (ablation)WildRefer · LifeReferAcc@0.25: 38.27 -> 38.89Improved
fuse language with point features first, then images (vs Vision First) (ablation)TFI · STReferAcc@0.5: 52.72 -> 54.97Improved
point-dominated rather than image-dominated fusion (ablation)TFI · STReferAcc@0.25: 52.25 -> 62.01Improved
fusion order variants (ablation)TFI · LifeReferAcc@0.25: Image Domin. 38.82 vs ours 38.82No clear effect
use 2 frames instead of 1 in DVE (ablation)DVE · STReferAcc@0.25: 59.38 -> 62.01Improved
increase frames from 2 to 3 or 4 (ablation)DVE · STReferAcc@0.25: 62.01 -> 59.94 (3 frames), 60.41 (4 frames)Hurt
replace DVE with feature concatenation across frames (ablation)DVE · STReferAcc@0.25: 62.01 -> 40.24Hurt
one-stage instead of two-stage pipeline for BUTD-DETRBUTD-DETR · LifeReferAcc@0.25: 30.81 -> 32.46Improved
WildRefer vs best prior method on STReferWildRefer · STReferAcc@0.5: 47.47 -> 54.97Improved

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