T4P: Test-Time Training of Trajectory Prediction via Masked Autoencoder and Actor-specific Token Memory

Documenti analoghi

• Multi-agent Long-term 3D Human Pose Forecasting via Interaction-aware Trajectory Conditioning
  di: Jeong, Jaewoo, et al.
  Pubblicazione: (2024)
• Multi-modal Knowledge Distillation-based Human Trajectory Forecasting
  di: Jeong, Jaewoo, et al.
  Pubblicazione: (2025)
• Interaction-Merged Motion Planning: Effectively Leveraging Diverse Motion Datasets for Robust Planning
  di: Lee, Giwon, et al.
  Pubblicazione: (2025)
• Improving Black-Box Generative Attacks via Generator Semantic Consistency
  di: Jeong, Jongoh, et al.
  Pubblicazione: (2025)
• Non-differentiable Reward Optimization for Diffusion-based Autonomous Motion Planning
  di: Lee, Giwon, et al.
  Pubblicazione: (2025)