AI Paper Atlas

Research map / Point clouds / Shape and surface completion

TouchDream: 3D Object Completion through Imagined Touch

Yuanbo Wang, Xinning Wang, Zhaoxuan Zhang, Changlong Wang, Qianchen Xia, Xiaopeng Wei, Xin Yang

CVPR 2026 · Point clouds › Shape and surface completion · approach: Point cloud completion networks

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 generated touch guidance instead of symmetry guidance (ablation)SymmCompletion-style refinement · PCNL1 CD avg (x10^3): 6.28 -> 6.05Improved
combine symmetry guidance with touch guidance (ablation)TouchDream · PCNL1 CD avg (x10^3): 6.05 -> 6.02No clear effect
use simulated ground truth touch instead of generated touch (ablation)TouchDream · PCNL1 CD avg (x10^3): 6.05 -> 5.93Improved
use skeleton points from ground truth instead of generated touch (ablation)TouchDream · PCNL1 CD avg (x10^3): 6.05 -> 6.04No clear effect
add test-time pose noise (R/T 0.02/0.01) without augmentation (ablation)TouchDream · PCNL1 CD avg (x10^3): 6.053 -> 6.183Hurt
add pose-noise data augmentation during training (ablation)TouchDream · PCN with pose noise R/T 0.02/0.01L1 CD avg (x10^3): 6.183 -> 6.087Improved
add imagined touch guidance to coarse completionSymmCompletion · ShapeNet-34 unseen categoriesL2 CD avg (x10^3): 0.97 -> 0.66Improved
add imagined touch guidance to coarse completionSymmCompletion · KITTI (trained on ShapeNet Car)FD / MMD: FD 2.54 -> 1.43; MMD 1.72 -> 0.71Improved
generate a touch latent per pose with diffusionTouchDream · PCN evaluation costevaluation time: 6.7 hours for 1,200 shapesHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 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