AI Paper Atlas

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

Touch and Go: Learning from Human-Collected Vision and Touch

Fengyu Yang, Chenyang Ma, Jiacheng Zhang, Jing Zhu, Wenzhen Yuan, Andrew Owens

NeurIPS 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Multimodal and affective interaction modeling · 20 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 visual input to tactile history for future touch prediction (Geng et al.) (ablation)Geng et al. video prediction · future touch prediction, skip 5 framesL1: 0.691 -> 0.663Improved
add visual input to tactile history (Geng et al.) (ablation)Geng et al. video prediction · future touch prediction, skip 3 framesLPIPS: 0.103 -> 0.091Improved
add visual input to tactile history (SVG) (ablation)SVG video prediction · future touch prediction, skip 5 framesL1: 0.807 -> 0.762Improved
increase fraction of material classes in pretraining from 25% to 100% (ablation)visuo-tactile CMC · Touch and Go material classificationaccuracy (%): 25.7 -> 54.7Improved
increase fraction of material classes in pretraining from 25% to 100% (ablation)visuo-tactile CMC · robot graspingaccuracy (%): 62.3 -> 78.1Improved
pretrain visuo-tactile CMC on Touch and Go instead of supervised ImageNet featurestactile representation · robot graspingaccuracy (%): 73.0 -> 78.1Improved
pretrain on Touch and Go instead of Object Folder 2.0visuo-tactile CMC · Touch and Go material classificationaccuracy (%): 36.2 -> 54.7Improved
use CUT-based stylization instead of CycleGAN (ablation)tactile-driven image stylization · tactile-driven stylizationCMC similarity: 0.178 -> 0.197Improved

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