---
language:
- en
license: cc-by-nc-sa-4.0
pipeline_tag: other
---

# MYRIAD (Envisioning the Future, One Step at a Time)

[![Project Page](https://img.shields.io/badge/Project-Page-blue)](https://compvis.github.io/myriad)
[![Paper](https://img.shields.io/badge/arXiv-paper-b31b1b)](https://arxiv.org/abs/2604.09527)
[![Paper](https://img.shields.io/badge/Huggingface-Papers-yellow)](https://huggingface.co/papers/2604.09527)
[![GitHub](https://img.shields.io/badge/GitHub-Code-black)](https://github.com/CompVis/flow-poke-transformer)
[![OWM-95](https://img.shields.io/badge/HuggingFace-Dataset-yellow)](https://huggingface.co/datasets/CompVis/owm-95)
[![MYRIAD-Physics Benchmark](https://img.shields.io/badge/Related-MYRIAD--physics-green)](https://huggingface.co/datasets/CompVis/myriad-physics)


MYRIAD (Motion hYpothesis Reasoning via Iterative Autoregressive Diffusion) is an autoregressive diffusion model that predicts open-set future scene dynamics as step-wise inference over sparse point trajectories. Starting from a single image, it can efficiently explore thousands of plausible future outcomes, maintaining physical plausibility.

## Paper and Abstract

The MYRIAD model was presented in the paper [Envisioning the Future, One Step at a Time](https://arxiv.org/abs/2604.09527).

From a single image, MYRIAD predicts distributions over sparse point trajectories autoregressively. This allows the model to predict consistent futures in open-set environments and plan actions by exploring a large number of counterfactual interactions.

![MYRIAD predicts distributions over potential motion auto-regressively](https://compvis.github.io/myriad/static/images/paper-svg/teaser-qualitative.svg)
*From a single image, our model envisions diverse, physically consistent futures by predicting sparse point trajectories step by step.*

![Sampling from MYRIAD enables planning-by-search](https://compvis.github.io/myriad/static/images/paper-svg/teaser-billiards.svg)
*Its efficiency enables exploring thousands of counterfactual rollouts directly in motion space - here illustrated for billiards planning, where candidate shots are evaluated by simulating many possible outcomes.*

## Usage

For programmatic use, the simplest way to use MYRIAD is via `torch.hub`:

```python
import torch

# Load the open-set model
myriad_openset = torch.hub.load("CompVis/myriad", "myriad_openset")

# Load the billiard-specific model
myriad_billiard = torch.hub.load("CompVis/myriad", "myriad_billiard")
```

If you wish to integrate MYRIAD in your own codebase, you can copy `model.py` and `dinov3.py` from the [GitHub repository](https://github.com/CompVis/flow-poke-transformer).
The `MyriadStepByStep` class contains a `predict_simulate` method for unrolling trajectories and a low-level `forward` method to predict distributions for previously observed trajectories.

## Citation

If you find our model or code useful, please cite our paper:

```bibtex
@inproceedings{baumann2026envisioning,
  title={Envisioning the Future, One Step at a Time},
  author={Baumann, Stefan Andreas and Wiese, Jannik and Martorella, Tommaso and Kalayeh, Mahdi M. and Ommer, Bjorn},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2026}
}
```