Guardado en:
Detalles Bibliográficos
Autores principales: Xia, Zhongyi, Wu, Tianzhao
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2409.01133
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912010968498176
author Xia, Zhongyi
Wu, Tianzhao
author_facet Xia, Zhongyi
Wu, Tianzhao
contents Monocular depth estimation is a critical function in computer vision applications. This paper shows that large language models (LLMs) can effectively interpret depth with minimal supervision, using efficient resource utilization and a consistent neural network architecture. We introduce LLM-MDE, a multimodal framework that deciphers depth through language comprehension. Specifically, LLM-MDE employs two main strategies to enhance the pretrained LLM's capability for depth estimation: cross-modal reprogramming and an adaptive prompt estimation module. These strategies align vision representations with text prototypes and automatically generate prompts based on monocular images, respectively. Comprehensive experiments on real-world MDE datasets confirm the effectiveness and superiority of LLM-MDE, which excels in few-/zero-shot tasks while minimizing resource use. The source code is available.
format Preprint
id arxiv_https___arxiv_org_abs_2409_01133
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Large Language Models Can Understanding Depth from Monocular Images
Xia, Zhongyi
Wu, Tianzhao
Computer Vision and Pattern Recognition
Artificial Intelligence
Monocular depth estimation is a critical function in computer vision applications. This paper shows that large language models (LLMs) can effectively interpret depth with minimal supervision, using efficient resource utilization and a consistent neural network architecture. We introduce LLM-MDE, a multimodal framework that deciphers depth through language comprehension. Specifically, LLM-MDE employs two main strategies to enhance the pretrained LLM's capability for depth estimation: cross-modal reprogramming and an adaptive prompt estimation module. These strategies align vision representations with text prototypes and automatically generate prompts based on monocular images, respectively. Comprehensive experiments on real-world MDE datasets confirm the effectiveness and superiority of LLM-MDE, which excels in few-/zero-shot tasks while minimizing resource use. The source code is available.
title Large Language Models Can Understanding Depth from Monocular Images
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2409.01133