NavDreamer: Video Models as Zero-Shot 3D Navigators
cs.RO
Submitted: 2026-02-10
Updated: 2026-09-21
Comments: Work in the progress. 22 pages, 15 figures
Project page: https://xinjiu612.github.io/NavDreamer
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Gemini: A Family of Highly Capable Multimodal Models
- Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond
- GPT-4 Technical Report
- RT-1: Robotics Transformer for Real-World Control at Scale
- OpenVLA: An Open-Source Vision-Language-Action Model
- Scaling Laws for Neural Language Models
- Scaling Laws for Transfer
- Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware
- UAV-Flow Colosseo: A Real-World Benchmark for Flying-on-a-Word UAV Imitation Learning
- Data Scaling Laws in Imitation Learning for Robotic Manipulation
- RH20T: A Comprehensive Robotic Dataset for Learning Diverse Skills in One-Shot
- Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots
- TrackVLA: Embodied Visual Tracking in the Wild
- VIRAL: Visual Sim-to-Real at Scale for Humanoid Loco-Manipulation
- Flying in Clutter on Monocular RGB by Learning in 3D Radiance Fields with Domain Adaptation
- Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning
- Video as the New Language for Real-World Decision Making
- Wan: Open and Advanced Large-Scale Video Generative Models
- HunyuanVideo 1.5 Technical Report
- World Simulation with Video Foundation Models for Physical AI
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies
- Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving