Guardado en:
Detalles Bibliográficos
Autores principales: Liu, Yong, Xu, Ruihao, Tang, Yansong
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2409.19569
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909328888299520
author Liu, Yong
Xu, Ruihao
Tang, Yansong
author_facet Liu, Yong
Xu, Ruihao
Tang, Yansong
contents This paper focuses on the Referring Image Segmentation (RIS) task, which aims to segment objects from an image based on a given language description. The critical problem of RIS is achieving fine-grained alignment between different modalities to recognize and segment the target object. Recent advances using the attention mechanism for cross-modal interaction have achieved excellent progress. However, current methods tend to lack explicit principles of interaction design as guidelines, leading to inadequate cross-modal comprehension. Additionally, most previous works use a single-modal mask decoder for prediction, losing the advantage of full cross-modal alignment. To address these challenges, we present a Fully Aligned Network (FAN) that follows four cross-modal interaction principles. Under the guidance of reasonable rules, our FAN achieves state-of-the-art performance on the prevalent RIS benchmarks (RefCOCO, RefCOCO+, G-Ref) with a simple architecture.
format Preprint
id arxiv_https___arxiv_org_abs_2409_19569
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Fully Aligned Network for Referring Image Segmentation
Liu, Yong
Xu, Ruihao
Tang, Yansong
Computer Vision and Pattern Recognition
This paper focuses on the Referring Image Segmentation (RIS) task, which aims to segment objects from an image based on a given language description. The critical problem of RIS is achieving fine-grained alignment between different modalities to recognize and segment the target object. Recent advances using the attention mechanism for cross-modal interaction have achieved excellent progress. However, current methods tend to lack explicit principles of interaction design as guidelines, leading to inadequate cross-modal comprehension. Additionally, most previous works use a single-modal mask decoder for prediction, losing the advantage of full cross-modal alignment. To address these challenges, we present a Fully Aligned Network (FAN) that follows four cross-modal interaction principles. Under the guidance of reasonable rules, our FAN achieves state-of-the-art performance on the prevalent RIS benchmarks (RefCOCO, RefCOCO+, G-Ref) with a simple architecture.
title Fully Aligned Network for Referring Image Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.19569