---
license: gpl-3.0
---

# DEVA: Tracking Anything with Decoupled Video Segmentation

![titlecard](https://imgur.com/lw15BGH.png)

[Ho Kei Cheng](https://hkchengrex.github.io/), [Seoung Wug Oh](https://sites.google.com/view/seoungwugoh/), [Brian Price](https://www.brianpricephd.com/), [Alexander Schwing](https://www.alexander-schwing.de/), [Joon-Young Lee](https://joonyoung-cv.github.io/)

University of Illinois Urbana-Champaign and Adobe

ICCV 2023

[[arXiV (coming soon)]]() [[PDF]](https://drive.google.com/file/d/1lAgg-j8d6EH1XYUz9htDaZDh4pxuIslb) [[Project Page]](https://hkchengrex.github.io/Tracking-Anything-with-DEVA/)

## Highlights
1. Provide long-term, open-vocabulary video segmentation with text-prompts out-of-the-box.
2. Fairly easy to **integrate your own image model**! Wouldn't you or your reviewers be interested in seeing examples where your image model also works well on videos :smirk:? No finetuning is needed!

## Abstract

We develop a decoupled video segmentation approach (**DEVA**), composed of task-specific image-level segmentation and class/task-agnostic bi-directional temporal propagation.
Due to this design, we only need an image-level model for the target task and a universal temporal propagation model which is trained once and generalizes across tasks.
To effectively combine these two modules, we propose a (semi-)online fusion of segmentation hypotheses from different frames to generate a coherent segmentation.
We show that this decoupled formulation compares favorably to end-to-end approaches in several tasks, most notably in large-vocabulary video panoptic segmentation and open-world video segmentation.