Guardado en:
Detalles Bibliográficos
Autores principales: Luo, Liqin, Chen, Guangyao, Zheng, Xiawu, Dai, Yongxing, Zou, Yixiong, Tian, Yonghong
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2511.19516
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911288088592384
author Luo, Liqin
Chen, Guangyao
Zheng, Xiawu
Dai, Yongxing
Zou, Yixiong
Tian, Yonghong
author_facet Luo, Liqin
Chen, Guangyao
Zheng, Xiawu
Dai, Yongxing
Zou, Yixiong
Tian, Yonghong
contents Visual grounding, the task of linking textual queries to specific regions within images, plays a pivotal role in vision-language integration. Existing methods typically rely on extensive task-specific annotations and fine-tuning, limiting their ability to generalize effectively to novel or out-of-distribution scenarios. To address these limitations, we introduce GroundingAgent, a novel agentic visual grounding framework that operates without any task-specific fine-tuning. GroundingAgent employs a structured, iterative reasoning mechanism that integrates pretrained open-vocabulary object detectors, multimodal large language models (MLLMs), and large language models (LLMs) to progressively refine candidate regions through joint semantic and spatial analyses. Remarkably, GroundingAgent achieves an average zero-shot grounding accuracy of 65.1 % on widely-used benchmarks (RefCOCO, RefCOCO+, RefCOCOg), entirely without fine-tuning. Furthermore, by substituting MLLM-generated captions with the original query texts, the accuracy at the selection stage alone reaches approximately 90 %, closely matching supervised performance and underscoring the critical role of LLM reasoning capabilities. GroundingAgent also offers strong interpretability, transparently illustrating each reasoning step and providing clear insights into its decision-making process.
format Preprint
id arxiv_https___arxiv_org_abs_2511_19516
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Connecting the Dots: Training-Free Visual Grounding via Agentic Reasoning
Luo, Liqin
Chen, Guangyao
Zheng, Xiawu
Dai, Yongxing
Zou, Yixiong
Tian, Yonghong
Computer Vision and Pattern Recognition
Visual grounding, the task of linking textual queries to specific regions within images, plays a pivotal role in vision-language integration. Existing methods typically rely on extensive task-specific annotations and fine-tuning, limiting their ability to generalize effectively to novel or out-of-distribution scenarios. To address these limitations, we introduce GroundingAgent, a novel agentic visual grounding framework that operates without any task-specific fine-tuning. GroundingAgent employs a structured, iterative reasoning mechanism that integrates pretrained open-vocabulary object detectors, multimodal large language models (MLLMs), and large language models (LLMs) to progressively refine candidate regions through joint semantic and spatial analyses. Remarkably, GroundingAgent achieves an average zero-shot grounding accuracy of 65.1 % on widely-used benchmarks (RefCOCO, RefCOCO+, RefCOCOg), entirely without fine-tuning. Furthermore, by substituting MLLM-generated captions with the original query texts, the accuracy at the selection stage alone reaches approximately 90 %, closely matching supervised performance and underscoring the critical role of LLM reasoning capabilities. GroundingAgent also offers strong interpretability, transparently illustrating each reasoning step and providing clear insights into its decision-making process.
title Connecting the Dots: Training-Free Visual Grounding via Agentic Reasoning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.19516