ReSTR: Convolution-free Referring Image Segmentation Using Transformers
- Title
- ReSTR: Convolution-free Referring Image Segmentation Using Transformers
- Authors
- 김남엽; 김동원; 곽수하; Lan, Cuiling; Zeng, Wenjun
- Date Issued
- 2022-06-22
- Publisher
- IEEE Computer Society
- Abstract
- Referring image segmentation is an advanced semantic segmentation task where target is not a predefined class but is described in natural language. Most of existing methods for this task rely heavily on convolutional neural networks, which however have trouble capturing long-range dependencies between entities in the language expression and are not flexible enough for modeling interactions between the two different modalities. To address these issues, we present the first convolution-free model for referring image segmentation using transformers, dubbed ReSTR. Since it extracts features of both modalities through transformer encoders, it can capture long-range dependencies between entities within each modality. Also, ReSTR fuses features of the two modalities by a self-attention encoder, which enables flexible and adaptive interactions between the two modalities in the fusion process. The fused features are fed to a segmentation module, which works adaptively according to the image and language expression in hand. ReSTR is evaluated and compared with previous work on all public benchmarks, where it outperforms all existing models.
- URI
- https://oasis.postech.ac.kr/handle/2014.oak/122826
- Article Type
- Conference
- Citation
- 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022, page. 18124 - 18133, 2022-06-22
- Files in This Item:
- There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.