# SPW-Nav streams 2K 360° video from language instructions

> Researchers introduce SPW-Nav, a model that generates one‑minute 2K panoramic video in real time from a single image, following spoken navigation cues.

Oossa · 2026-10-08 · https://oossa.com/en/spw-nav-streams-2k-360-video-from-language-instructions

The team led by Ziqi Cai released SPW-Nav, a streaming panoramic world model that turns language directions into live 360‑degree video. It works from one starting panorama and can follow instructions like “turn left” or “move forward” while keeping video quality high. The system uses spherical rotation and pose‑aligned conditioning to avoid drift over long streams. They also released SPW-NavSet, a dataset of panoramic videos with verified camera trajectories and instructions.

## The facts

- Paper posted on arXiv on 2026-10-08
- Generates up to one minute of 2K 360° video in real time

## Why it matters

It could let VR apps create on‑the‑fly tours from a single photo, reducing the need for pre‑recorded footage.

## Sources & references

1. [SPW-Nav: A Streaming Panoramic World Model for Language-Guided Navigation](https://arxiv.org/abs/2610.08941) – arXiv cs.CV, 2026-10-08

Last updated: 2026-10-08