mlsanityData to 14 Oct 2026

Label-Efficient Grasp Joint Prediction with Point-JEPA

arXiv 2509.13349 ↗First posted 13 Sep 2025cs.RO cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

We study whether 3D self-supervised pretraining with Point--JEPA enables label-efficient grasp joint-angle prediction. Meshes are sampled to point clouds and tokenized; a ShapeNet-pretrained Point--JEPA encoder feeds a K=5 multi-hypothesis head trained with winner-takes-all and evaluated by top--logit selection. On a multi-finger hand dataset with strict object-level splits, Point--JEPA improves top--logit RMSE and Coverage@15circ in low-label regimes (e.g., 26% lower RMSE at 25% data) and reaches parity at full supervision, suggesting JEPA-style pretraining is a practical lever for data-efficient grasp learning.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 15 of its 17 references that are in the map. First posted on arXiv 13 Sep 2025.

Builds on

15 of 17 references.

References · 15

  1. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  2. 2023Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecturecited by 1,039
  3. 2022data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Languagecited by 727
  4. 2021Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modelingcited by 653
  5. 2022Masked Autoencoders for Point Cloud Self-supervised Learningcited by 459
  6. 2017Dex-Net 2.0: Deep Learning to Plan Robust Grasps with Synthetic Point Clouds and Analytic Grasp Metricscited by 264
  7. 2019CoverNet: Multimodal Behavior Prediction using Trajectory Setscited by 259
  8. 2021Contact-GraspNet: Efficient 6-DoF Grasp Generation in Cluttered Scenescited by 218
  9. 2016Learning in an Uncertain World: Representing Ambiguity Through Multiple Hypothesescited by 97
  10. 2022Masked Autoencoder for Self-Supervised Pre-training on Lidar Point Cloudscited by 39
  11. 2016High precision grasp pose detection in dense cluttercited by 34
  12. 2024Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloudcited by 30
3 more
  1. 2023Point2Vec for Self-Supervised Representation Learning on Point Cloudscited by 28
  2. 20243D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learningcited by 14
  3. 2023Combining Shape Completion and Grasp Prediction for Fast and Versatile Grasping with a Multi-Fingered Handcited by 4

Built on by

Not cited yet.