Zapisane w:
Opis bibliograficzny
Główni autorzy: Xiao, Linhui, Yang, Xiaoshan, Peng, Fang, Wang, Yaowei, Xu, Changsheng
Format: Preprint
Wydane: 2024
Hasła przedmiotowe:
Dostęp online:https://arxiv.org/abs/2410.08021
Etykiety: Dodaj etykietę
Nie ma etykietki, Dołącz pierwszą etykiete!
_version_ 1866914987940773888
author Xiao, Linhui
Yang, Xiaoshan
Peng, Fang
Wang, Yaowei
Xu, Changsheng
author_facet Xiao, Linhui
Yang, Xiaoshan
Peng, Fang
Wang, Yaowei
Xu, Changsheng
contents Constrained by the separate encoding of vision and language, existing grounding and referring segmentation works heavily rely on bulky Transformer-based fusion en-/decoders and a variety of early-stage interaction technologies. Simultaneously, the current mask visual language modeling (MVLM) fails to capture the nuanced referential relationship between image-text in referring tasks. In this paper, we propose OneRef, a minimalist referring framework built on the modality-shared one-tower transformer that unifies the visual and linguistic feature spaces. To modeling the referential relationship, we introduce a novel MVLM paradigm called Mask Referring Modeling (MRefM), which encompasses both referring-aware mask image modeling and referring-aware mask language modeling. Both modules not only reconstruct modality-related content but also cross-modal referring content. Within MRefM, we propose a referring-aware dynamic image masking strategy that is aware of the referred region rather than relying on fixed ratios or generic random masking schemes. By leveraging the unified visual language feature space and incorporating MRefM's ability to model the referential relations, our approach enables direct regression of the referring results without resorting to various complex techniques. Our method consistently surpasses existing approaches and achieves SoTA performance on both grounding and segmentation tasks, providing valuable insights for future research. Our code and models are available at https://github.com/linhuixiao/OneRef.
format Preprint
id arxiv_https___arxiv_org_abs_2410_08021
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle OneRef: Unified One-tower Expression Grounding and Segmentation with Mask Referring Modeling
Xiao, Linhui
Yang, Xiaoshan
Peng, Fang
Wang, Yaowei
Xu, Changsheng
Computer Vision and Pattern Recognition
Constrained by the separate encoding of vision and language, existing grounding and referring segmentation works heavily rely on bulky Transformer-based fusion en-/decoders and a variety of early-stage interaction technologies. Simultaneously, the current mask visual language modeling (MVLM) fails to capture the nuanced referential relationship between image-text in referring tasks. In this paper, we propose OneRef, a minimalist referring framework built on the modality-shared one-tower transformer that unifies the visual and linguistic feature spaces. To modeling the referential relationship, we introduce a novel MVLM paradigm called Mask Referring Modeling (MRefM), which encompasses both referring-aware mask image modeling and referring-aware mask language modeling. Both modules not only reconstruct modality-related content but also cross-modal referring content. Within MRefM, we propose a referring-aware dynamic image masking strategy that is aware of the referred region rather than relying on fixed ratios or generic random masking schemes. By leveraging the unified visual language feature space and incorporating MRefM's ability to model the referential relations, our approach enables direct regression of the referring results without resorting to various complex techniques. Our method consistently surpasses existing approaches and achieves SoTA performance on both grounding and segmentation tasks, providing valuable insights for future research. Our code and models are available at https://github.com/linhuixiao/OneRef.
title OneRef: Unified One-tower Expression Grounding and Segmentation with Mask Referring Modeling
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.08021