Does MoCam work in the EU?

Unverified. No explicit evidence found. It is a research project page with no public code, weights or service. Last verified 14 Nov 2026.

How much does MoCam cost?

Pricing unknown. No pricing or public release found; code and models reported as coming soon.

Where was MoCam shown?

In 1 creator video by AI Search. Most recently by AI Search on 17 May 2026, at 34:50 in "Real gundams, top 3D generator, open-source world models, ChatGPT updates, new TTS: AI NEWS".

MoCam

Website Paper

Research method for re-rendering an existing video along a new camera trajectory while preserving subject motion.

UnverifiedVideo GenerationModel

Updated 14 Nov 2026EU status checked 14 Nov 2026Last shown 17 May 2026

MoCam at a glance

What is MoCam?: MoCam is a model. Research method for re-rendering an existing video along a new camera trajectory while preserving subject motion.
Does MoCam work in the EU?: Unverified. No explicit evidence found. It is a research project page with no public code, weights or service. Last verified 14 Nov 2026.
How much does MoCam cost?: Pricing unknown. No pricing or public release found; code and models reported as coming soon.
Where was MoCam shown?: In 1 creator video by AI Search. Most recently by AI Search on 17 May 2026, at 34:50 in “Real gundams, top 3D generator, open-source world models, ChatGPT updates, new TTS: AI NEWS”.

Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.

About

MoCam is a research method for video re-camera and novel view synthesis. Given an existing video, it generates a new version with a different camera path while keeping the subject motion. It targets both static and dynamic scenes.

The method uses "structured denoising dynamics" in a diffusion process. Generation moves from initial geometric alignment to final appearance refinement. According to the project page, this lets it work with highly incomplete or distorted geometric priors. The demos cover large-movement and complex trajectories, dolly zoom, zoom in/out and bullet-time effects.

The project page lists the paper as arXiv:2605.12119, with authors Haofeng Liu, Yang Zhou, Ziheng Wang, Zhengbo Xu, Zhan Peng, Jie Ma, Jun Liang, Shengfeng He and Jing Li. A video creator reported that code and models were "coming soon". The provided material shows no public code, weights, hosted demo or product, and no pricing or licence information.

Key features