Olympus: A Universal Task Router for Computer Vision Tasks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lin, Yuanze, Li, Yunsheng, Chen, Dongdong, Xu, Weijian, Clark, Ronald, Torr, Philip H. S.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908295129726976
author Lin, Yuanze
Li, Yunsheng
Chen, Dongdong
Xu, Weijian
Clark, Ronald
Torr, Philip H. S.
author_facet Lin, Yuanze
Li, Yunsheng
Chen, Dongdong
Xu, Weijian
Clark, Ronald
Torr, Philip H. S.
contents We introduce Olympus, a new approach that transforms Multimodal Large Language Models (MLLMs) into a unified framework capable of handling a wide array of computer vision tasks. Utilizing a controller MLLM, Olympus delegates over 20 specialized tasks across images, videos, and 3D objects to dedicated modules. This instruction-based routing enables complex workflows through chained actions without the need for training heavy generative models. Olympus easily integrates with existing MLLMs, expanding their capabilities with comparable performance. Experimental results demonstrate that Olympus achieves an average routing accuracy of 94.75% across 20 tasks and precision of 91.82% in chained action scenarios, showcasing its effectiveness as a universal task router that can solve a diverse range of computer vision tasks. Project page: http://yuanze-lin.me/Olympus_page/
format Preprint
id arxiv_https___arxiv_org_abs_2412_09612
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Olympus: A Universal Task Router for Computer Vision Tasks
Lin, Yuanze
Li, Yunsheng
Chen, Dongdong
Xu, Weijian
Clark, Ronald
Torr, Philip H. S.
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
We introduce Olympus, a new approach that transforms Multimodal Large Language Models (MLLMs) into a unified framework capable of handling a wide array of computer vision tasks. Utilizing a controller MLLM, Olympus delegates over 20 specialized tasks across images, videos, and 3D objects to dedicated modules. This instruction-based routing enables complex workflows through chained actions without the need for training heavy generative models. Olympus easily integrates with existing MLLMs, expanding their capabilities with comparable performance. Experimental results demonstrate that Olympus achieves an average routing accuracy of 94.75% across 20 tasks and precision of 91.82% in chained action scenarios, showcasing its effectiveness as a universal task router that can solve a diverse range of computer vision tasks. Project page: http://yuanze-lin.me/Olympus_page/
title Olympus: A Universal Task Router for Computer Vision Tasks
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2412.09612