AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

3D-VisTA: Pre-trained Transformer for 3D Vision and Text Alignment

Ziyu Zhu, Xiaojian Ma, Yixin Chen, Zhidong Deng, Siyuan Huang, Qing Li

ICCV 2023 · Vision-language and multimodal › Human and embodied modeling · approach: Region-level vision-language alignment · 162 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
pre-train on ScanScribe before fine-tuning (ablation)3D-VisTA · Nr3Doverall accuracy (%): 57.5 -> 64.2Improved
pre-train on ScanScribe before fine-tuning (ablation)3D-VisTA · ScanRefer (Mask3D)overall acc@0.5 (%): 41.5 -> 45.8Improved
pre-train on ScanScribe before fine-tuning (ablation)3D-VisTA · Scan2CapCIDEr@0.25: 66.8 -> 71.0; BLEU-4 36.6 -> 36.5Mixed
pre-train on ScanScribe before fine-tuning (ablation)3D-VisTA · SQA3Daverage accuracy (%): 46.7 -> 48.5Improved
add pre-training on 30-40% of annotations vs scratch with full data3D-VisTA · ScanRefer / ScanQAaccuracyImproved
increase fusion Transformer depth beyond 4 layers (ablation)3D-VisTA · ScanRefer VGgrounding accuracy (%): 57.4 -> 56.3 (8 layers)No clear effect
pre-train with MLM only (ablation)3D-VisTA · ScanRefer VGgrounding accuracy (%): 52.0 -> 51.5Hurt
add masked object modeling (MOM) to MLM (ablation)3D-VisTA pre-training · ScanRefer VGgrounding accuracy (%): 51.5 -> 57.1Improved
add scene-text matching (STM) to MLM + MOM (ablation)3D-VisTA pre-training · ScanQA EM@1EM@1 (%): 22.5 -> 23.8Improved
add 3R-Scan and Objaverse object replacement to ScanNet pre-training data (ablation)3D-VisTA pre-training · ScanRefer VGgrounding accuracy (%): 54.6 -> 57.4Improved
use Mask3D instead of PointGroup proposals (ablation)3D-VisTA · ScanReferoverall acc@0.5 (%): 37.3 -> 45.8Improved

Later papers that compared with this method

7 later accepted papers report 3D-VisTA as a baseline in their results tables; 6 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