AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

RoboPearls: Editable Video Simulation for Robot Manipulation

Tang Tao, Likui Zhang, Youpeng Wen, Kaidong Zhang, Jia-Wang Bian, Xia Zhou, Tianyi Yan, Kun Zhan, Peng Jia, Hefeng Wu, Liang Lin, Xiaodan Liang

ICCV 2025 · Vision-language and multimodal › Robot action and navigation · approach: Memory and subgoal-based manipulation · 1 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
replace 2D IP2P editing with RoboPearls 3D simulation (ablation)RVT · RLBench Stack Cupssuccess rate: 18.4 -> 37.7Improved
replace 2D IP2P editing with RoboPearls 3D simulation (ablation)RVT · RLBench Put in Cupboardsuccess rate: 44.8 -> 55.5Improved
add VLM failure analysis to RoboPearls (ablation)RoboPearls · RLBench Stack Cupssuccess rate: 24.7 -> 37.7Improved
add VLM failure analysis to RoboPearls (ablation)RoboPearls · RLBench Put in Cupboardsuccess rate: 45.0 -> 55.5Improved
add VLM failure analysis to RoboPearls (ablation)RoboPearls · RLBench Insert Pegsuccess rate: 16.5 -> 17.1No clear effect
add IP2P edited data to RVT (ablation)RVT · RLBench Insert Pegsuccess rate: 11.0 -> 10.7No clear effect
train with RoboPearls simulation dataRVT · COLOSSEUM averagesuccess rate: 51.7 -> 69.2Improved
train with RoboPearls simulation dataRVT-2 · COLOSSEUM averagesuccess rate: 64.6 -> 75.4Improved
train with RoboPearls simulation dataRVT-2 · COLOSSEUM all perturbationssuccess rate: 36.1 -> 54.7Improved
train with RoboPearls simulation dataRVT · RLBench averagesuccess rate: 62.4 -> 68.0Improved
train with RoboPearls simulation dataSAM2Act · RLBench averagesuccess rate: 83.8 -> 88.5Improved
train RDT with RoboPearls simulation dataRDT · real robot Place in (unseen)successes out of 20: 1 / 20 -> 12 / 20Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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