Saved in:
Bibliographic Details
Main Authors: Zhu, Guiying, Yang, Bowen, Zhuang, Yin, Zhang, Tong, Wang, Guanqun, Che, Zhihao, Chen, He, Li, Lianlin
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.11910
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908778503340032
author Zhu, Guiying
Yang, Bowen
Zhuang, Yin
Zhang, Tong
Wang, Guanqun
Che, Zhihao
Chen, He
Li, Lianlin
author_facet Zhu, Guiying
Yang, Bowen
Zhuang, Yin
Zhang, Tong
Wang, Guanqun
Che, Zhihao
Chen, He
Li, Lianlin
contents Open-Vocabulary Object Detection (OVOD) aims to develop the capability to detect anything. Although myriads of large-scale pre-training efforts have built versatile foundation models that exhibit impressive zero-shot capabilities to facilitate OVOD, the necessity of creating a universal understanding for any object cognition according to already pretrained foundation models is usually overlooked. Therefore, in this paper, a training-free Guess What Vision Language Model, called GW-VLM, is proposed to form a universal understanding paradigm based on our carefully designed Multi-Scale Visual Language Searching (MS-VLS) coupled with Contextual Concept Prompt (CCP) for OVOD. This approach can engage a pre-trained Vision Language Model (VLM) and a Large Language Model (LLM) in the game of "guess what". Wherein, MS-VLS leverages multi-scale visual-language soft-alignment for VLM to generate snippets from the results of class-agnostic object detection, while CCP can form the concept of flow referring to MS-VLS and then make LLM understand snippets for OVOD. Finally, the extensive experiments are carried out on natural and remote sensing datasets, including COCO val, Pascal VOC, DIOR, and NWPU-10, and the results indicate that our proposed GW-VLM can achieve superior OVOD performance compared to the-state-of-the-art methods without any training step.
format Preprint
id arxiv_https___arxiv_org_abs_2601_11910
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Training-Free Guess What Vision Language Model from Snippets to Open-Vocabulary Object Detection
Zhu, Guiying
Yang, Bowen
Zhuang, Yin
Zhang, Tong
Wang, Guanqun
Che, Zhihao
Chen, He
Li, Lianlin
Computer Vision and Pattern Recognition
Open-Vocabulary Object Detection (OVOD) aims to develop the capability to detect anything. Although myriads of large-scale pre-training efforts have built versatile foundation models that exhibit impressive zero-shot capabilities to facilitate OVOD, the necessity of creating a universal understanding for any object cognition according to already pretrained foundation models is usually overlooked. Therefore, in this paper, a training-free Guess What Vision Language Model, called GW-VLM, is proposed to form a universal understanding paradigm based on our carefully designed Multi-Scale Visual Language Searching (MS-VLS) coupled with Contextual Concept Prompt (CCP) for OVOD. This approach can engage a pre-trained Vision Language Model (VLM) and a Large Language Model (LLM) in the game of "guess what". Wherein, MS-VLS leverages multi-scale visual-language soft-alignment for VLM to generate snippets from the results of class-agnostic object detection, while CCP can form the concept of flow referring to MS-VLS and then make LLM understand snippets for OVOD. Finally, the extensive experiments are carried out on natural and remote sensing datasets, including COCO val, Pascal VOC, DIOR, and NWPU-10, and the results indicate that our proposed GW-VLM can achieve superior OVOD performance compared to the-state-of-the-art methods without any training step.
title A Training-Free Guess What Vision Language Model from Snippets to Open-Vocabulary Object Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.11910