AI Paper Atlas

Research map / 3D vision / 3D generation

Ross3D: Reconstructive Visual Instruction Tuning with 3D-Awareness

Haochen Wang, Yucheng Zhao, Tiancai Wang, Haoqiang Fan, Xiangyu Zhang, Zhaoxiang Zhang

ICCV 2025 · 3D vision › 3D generation · approach: Learning 3D reconstruction from images · 49 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 vanilla reconstruction without 3D-awareness (ablation)Video-3D-LLM · SQA3DEM: 58.6 -> 58.8No clear effect
add cross-view reconstruction (ablation)Video-3D-LLM · ScanReferAcc@0.25: 58.1 -> 60.3Improved
add global-view reconstruction (ablation)Video-3D-LLM · SQA3DEM: 58.6 -> 61.6Improved
add both cross-view and global-view reconstruction with vanilla (ablation)Video-3D-LLM · ScanQACIDEr: 102.1 -> 107.0Improved
add both pretext tasks to point-based 3D-LLM (ablation)3D-LLM · SQA3DEM: 49.4 -> 55.0Improved
add global-view reconstruction to point-based 3D-LLM (ablation)3D-LLM · SQA3DEM: 49.4 -> 54.3Improved
add scene-level 3D point features at input instead of 3D-aware supervision (ablation)Video-3D-LLM · ScanReferAcc@0.25: 58.1 -> 57.8No clear effect
add L_3D on 50% unlabeled vision-only data (ablation)Ross3D with 50% text · ScanQACIDEr: 101.7 -> 103.2Improved
raise mask ratio from 0.25 to 0.75 (ablation)Ross3D · SQA3DEM: 63.0 -> 61.2Hurt
apply reconstruction every step (interval 1) instead of every 4 (ablation)Ross3D · ScanQACIDEr: 107.0 -> 104.8Hurt
filter black blanks from BEV images (ablation)Ross3D · ScanQACIDEr: 104.6 -> 107.0Improved
raise BEV resolution from 432x432 to 1024x1024 (ablation)Ross3D · ScanQACIDEr: 107.0 -> 106.5No clear effect

Later papers that compared with this method

2 later accepted papers report Ross3D as a baseline in their results tables; 2 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