ViTaPEs: Visuotactile Position Encodings for Cross-Modal Alignment in Multimodal Transformers

Documenti analoghi

• M2CURL: Sample-Efficient Multimodal Reinforcement Learning via Self-Supervised Representation Learning for Robotic Manipulation
  di: Lygerakis, Fotios, et al.
  Pubblicazione: (2024)
• Learning the RoPEs: Better 2D and 3D Position Encodings with STRING
  di: Schenck, Connor, et al.
  Pubblicazione: (2025)
• Restoring Vision in Adverse Weather Conditions with Patch-Based Denoising Diffusion Models
  di: Özdenizci, Ozan, et al.
  Pubblicazione: (2022)
• ViTa-Zero: Zero-shot Visuotactile Object 6D Pose Estimation
  di: Li, Hongyu, et al.
  Pubblicazione: (2025)
• Multimodal Visual-Tactile Representation Learning through Self-Supervised Contrastive Pre-Training
  di: Dave, Vedant, et al.
  Pubblicazione: (2024)