The N-Body Problem:

Parallel Execution from Single-Person Egocentric Video

Zhifan Zhu1, Yifei Huang2, Yoichi Sato2, Dima Damen1
1University of Bristol, 2The University of Tokyo
Arxiv

Starting from a single-person egocentric video,
we imagine how the task can be performed faster by N people.
A valid execution has to respect physical and causal constraints

Teaser

Humans can intuitively parallelise complex activities, but can a model learn this from observing a single person?
Given one egocentric video, we introduce the N-Body Problem: how N individuals, can hypothetically perform the same set of tasks observed in this video. The goal is to maximise speed-up, but naive assignment of video segments to individuals often violates real-world constraints, leading to physically impossible scenarios like two people using the same object or occupying the same space.

We formalise the N-Body Problem and propose a suite of metrics to evaluate both performance (speed-up, task coverage) and feasibility (spatial collisions, object conflicts and causal constraints).

We use structured prompting on Gemini Pro to reason about the 3D environment, object usage, and temporal dependencies to produce a viable parallel execution.

Interactive Demo (Move Between 1-Body [left] and 2-Body [right] Execution)

Single-Person Video

This has no parallel moment

Parallel Execution (N=2)

Video Demo

Citation


                @article{zhu2025nbody,
                    title = {The N-Body Problem: Parallel Execution from Single-Person Egocentric Video},
                    author = {Zhu, Zhifan and Huang, Yifei and Sato, Yoichi and Damen, Dima},
                    year = {2025},
                    eprint = {2512.11393},
                    archivePrefix = {arXiv},
                    primaryClass = {cs.CV}
                }