synthesizing-worlds-with-video-diffusion-models-is-often

Xingang Pan (@XingangP)

2025-04-18 | ❤️ 146 | 🔁 27

---

Synthesizing worlds with video diffusion models is often inconsistent — moving the camera back and forth leads to different scenes. We propose 🌐𝗪𝗼𝗿𝗹𝗱𝗠𝗲𝗺, a memory-based approach that ensures consistent world simulation without relying on explicit 3D reconstruction. https://x.com/XingangP/status/1913161004672004253/video/1

🔗 원본 링크

• https://x.com/XingangP/status/1913161004672004253/video/1

미디어

---

🔗 Related

Auto-generated - needs manual review

인용 트윗

Zeqi Xiao (@zeqi_xiao)

While recent works like Genie 2, The Matrix, and Navigation World Models explore video generative models as world simulators, world consistency remains underexplored. In this work, we propose 🌐WorldMem🌐, introducing a memory mechanism for long-term consistent world simulation. https://t.co/O3MTQxrLV6

원본 트윗

🎬 영상

Tags

domain-vision-3d domain-genai domain-ai-ml domain-simulation