StreamPETR offers efficient 3D object detection for autonomous driving by propagating sparse object queries temporally.
Accelerate post-training of end-to-end autonomous vehicle stacks with vector search and retrieval for large video datasets.
Reasoning vision language model (VLM) for physical AI and robotics.
Generates physics-aware video world states for physical AI development using text prompts and multiple spatial control inputs derived from real-world data or simulation.
End-to-end autonomous driving stack integrating perception, prediction, and planning with sparse scene representations for efficiency and safety.
