A Simple Baseline with Single-encoder for Referring Image Segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yu, Seonghoon, Jung, Ilchae, Han, Byeongju, Kim, Taeoh, Kim, Yunho, Wee, Dongyoon, Son, Jeany
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908409553485824
author Yu, Seonghoon
Jung, Ilchae
Han, Byeongju
Kim, Taeoh
Kim, Yunho
Wee, Dongyoon
Son, Jeany
author_facet Yu, Seonghoon
Jung, Ilchae
Han, Byeongju
Kim, Taeoh
Kim, Yunho
Wee, Dongyoon
Son, Jeany
contents Referring image segmentation (RIS) requires dense vision-language interactions between visual pixels and textual words to segment objects based on a given description. However, commonly adapted dual-encoders in RIS, e.g., Swin transformer and BERT (uni-modal encoders) or CLIP (a multi-modal dual-encoder), lack dense multi-modal interactions during pre-training, leading to a gap with a pixel-level RIS task. To bridge this gap, existing RIS methods often rely on multi-modal fusion modules that interact two encoders, but this approach leads to high computational costs. In this paper, we present a novel RIS method with a single-encoder, i.e., BEiT-3, maximizing the potential of shared self-attention across all framework components. This enables seamless interactions of two modalities from input to final prediction, producing granularly aligned multi-modal features. Furthermore, we propose lightweight yet effective decoder modules, a Shared FPN and a Shared Mask Decoder, which contribute to the high efficiency of our model. Our simple baseline with a single encoder achieves outstanding performances on the RIS benchmark datasets while maintaining computational efficiency, compared to the most recent SoTA methods based on dual-encoders.
format Preprint
id arxiv_https___arxiv_org_abs_2408_15521
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Simple Baseline with Single-encoder for Referring Image Segmentation
Yu, Seonghoon
Jung, Ilchae
Han, Byeongju
Kim, Taeoh
Kim, Yunho
Wee, Dongyoon
Son, Jeany
Computer Vision and Pattern Recognition
Multimedia
Referring image segmentation (RIS) requires dense vision-language interactions between visual pixels and textual words to segment objects based on a given description. However, commonly adapted dual-encoders in RIS, e.g., Swin transformer and BERT (uni-modal encoders) or CLIP (a multi-modal dual-encoder), lack dense multi-modal interactions during pre-training, leading to a gap with a pixel-level RIS task. To bridge this gap, existing RIS methods often rely on multi-modal fusion modules that interact two encoders, but this approach leads to high computational costs. In this paper, we present a novel RIS method with a single-encoder, i.e., BEiT-3, maximizing the potential of shared self-attention across all framework components. This enables seamless interactions of two modalities from input to final prediction, producing granularly aligned multi-modal features. Furthermore, we propose lightweight yet effective decoder modules, a Shared FPN and a Shared Mask Decoder, which contribute to the high efficiency of our model. Our simple baseline with a single encoder achieves outstanding performances on the RIS benchmark datasets while maintaining computational efficiency, compared to the most recent SoTA methods based on dual-encoders.
title A Simple Baseline with Single-encoder for Referring Image Segmentation
topic Computer Vision and Pattern Recognition
Multimedia
url https://arxiv.org/abs/2408.15521