MFNet: Multiclass Few-Shot Segmentation Network With Pixel-Wise Metric Learning

Zhang, Miao; Shi, Miaojing; Li, Li

doi:10.1109/tcsvt.2022.3193612

Cited by 20 publications

(3 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It can better discover co-occurring objects in two images through a cross-reference mechanism, helping to complete the task of few-shot segmentation. MFNet [35] is a novel multiplexed (class) encoding and decoding architecture. It successfully combines multi-scale query and multi-class support data into a single query support embedment.…”

Section: Related Workmentioning

confidence: 99%

A Novel Medical Decision-Making System Based on Multi-Scale Feature Enhancement for Small Samples

Qin

Gou

et al. 2023

Mathematics

View full text Add to dashboard Cite

The medical decision-making system is an advanced system for patients that can assist doctors in their medical work. Osteosarcoma is a primary malignant tumor of the bone, due to its specificity, such as its blurred borders, diverse tumor morphology, and inconsistent scales. Diagnosis is quite difficult, especially for developing countries, where medical resources are inadequate per capita and there is a lack of professionals, and the time spent in the diagnosis process may lead to a gradual deterioration of the disease. To address these, we discuss an osteosarcoma-assisted diagnosis system (OSADS) based on small samples with multi-scale feature enhancement that can assist doctors in performing preliminary automatic segmentation of osteosarcoma and reduce the workload. We proposed a multi-scale feature enhancement network (MFENet) based on few-shot learning in OSADS. Global and local feature information is extracted to effectively segment the boundaries of osteosarcoma by feeding the images into MFENet. Simultaneously, a prior mask is introduced into the network to help it maintain a certain accuracy range when segmenting different shapes and sizes, saving computational costs. In the experiments, we used 5000 osteosarcoma MRI images provided by Monash University for testing. The experiments show that our proposed method achieves 93.1% accuracy and has the highest comprehensive evaluation index compared with other methods.

show abstract

Section: Related Workmentioning

confidence: 99%

A Novel Medical Decision-Making System Based on Multi-Scale Feature Enhancement for Small Samples

Qin

Gou

et al. 2023

Mathematics

View full text Add to dashboard Cite

show abstract

“…This requires the model to understand the spatial context, boundaries, and appearance of objects despite the limited amount of labeled data. Existing methods can be categorized into two main groups: meta-learning and metric-based learning [12]- [15]. With meta-learning, the model can quickly adapt to new categories in several sample contexts, thus better handling unseen semantic categories.…”

Section: Introductionmentioning

confidence: 99%

Prototype Comparison Convolutional Networks for One-Shot Segmentation

Li,

Guo

et al. 2024

IEEE Access

View full text Add to dashboard Cite

In few-shot semantic segmentation (FSS), the key challenges are efficiently tuning the interaction between the support set and the query set and distinguishing between context, background, and interfering items. To address these challenges, we propose prototype comparison networks for oneshot segmentation (OPCN) to capture the details required for FSS. Specifically, we offer the Fusion Interaction Module (FIM) to improve the segmentation performance by capturing the correlation and semantic information between the support set and query set features. Subsequently, we propose the Feature Enhancement Module (FEM), which aims to enhance the information required in the support set and query set features while increasing the focus on critical details by reducing the weight of the background regions to provide a more targeted feature representation for subsequent query image segmentation. Then, we propose the Feature Refinement Module (FRM) to filter irrelevant background information in the features and specify the target location region. Finally, the Feature Matching Module (FM) generates the final segmentation mask for the query image. Extensive experiments on the PASCAL-5i and COCO-20i datasets show that our approach achieves excellent performance in the case of the one-shot setup.

show abstract

“…However, despite their effectiveness, pure textual information still cannot fully replicate a rich visual perceptual experience. To address this limitation, researchers have turned their attention to various visual language tasks, such as visual quizzing (Li et al; Zhang et al) [6,7], image and video caption generation (Chen F; Ghanimifard and Dobnik; Corniaet al) [8][9][10], and image-based question retrieval (Xin Yuan et al; Lu et al) [11,12]. In human conversational communication, images are crucial in compensating for information that cannot be accurately expressed through text alone.…”

Section: Introductionmentioning

confidence: 99%

Multimodal Dialogue Response Generation Based on Selective Attention and Gating Mechanisms

Yu,

Zhang,

Ding

et al. 2023

Preprint

View full text Add to dashboard Cite

Picture response is a crucial aspect in dialogue systems; however, existing systems predominantly focus on text information, often neglecting the use of image data or relying on large-scale language models, which leads to high training costs and slower generation speeds. To address these challenges, this paper introduces the MDSAGM (Selective Attention and Gating Mechanism)model, which comprises a Transformer-based Text Dialogue Response Generator, Text-to-Image Generation, and SAGM (Selective Attention and Gating Mechanism). The model is core lies in combining Selective Attention and Gating Mechanism to enhance generalization and improve model accuracy. The study further explores the contributions of image and text information in multimodal fusion. Compared to other large-scale training models, the proposed model demonstrates computational efficiency , fewer parameters, and shorter response times, making it more lightweight. This dialogue model facilitates the separation of multimodal dialogue parameters from the overall model, enabling better parameter fitting through pre-training with abundant plain text and text-image data. Extensive experiments show that the proposed method achieves promising results in both automatic and manual evaluations, generating information-rich text and image responses with higher accuracy while maintaining a lightweight structure.

show abstract

MFNet: Multiclass Few-Shot Segmentation Network With Pixel-Wise Metric Learning

Cited by 20 publications

References 58 publications

A Novel Medical Decision-Making System Based on Multi-Scale Feature Enhancement for Small Samples

A Novel Medical Decision-Making System Based on Multi-Scale Feature Enhancement for Small Samples

Prototype Comparison Convolutional Networks for One-Shot Segmentation

Multimodal Dialogue Response Generation Based on Selective Attention and Gating Mechanisms

Contact Info

Product

Resources

About