Fangfang Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Fangfang Wang has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Xi Li
Zhejiang University, Hangzhou, China
2 shared publications - Yifeng Chen
Department of Pathology, Quanzhou First Hospital Affiliated to Fujian Medical University, Quanzhou, China
2 shared publications - Canghong Jin
Zhejiang Provincial Engineering Research Center for Real-Time SmartTech in Urban Security Governance, Hangzhou City University, Hangzhou, China
1 shared publication - Chao-Ton Su
Department of Industrial Engineering and Engineering Management, National Tsing Hua University, Hsinchu, Taiwan
1 shared publication - Chengjie Wang
YouTu Lab, Tencent, Shanghai, China
1 shared publication Cuiling Wu
1 shared publication
Latest Publications
- Open Access
- 24 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (7) , 1-18
Hyperbolic Active Learning for Label-Efficient Action Segmentation
Recent advances in action segmentation have greatly enhanced our understanding of complex and dynamic scenes in video content. Despite these improvements, the field continues to face persistent challenges, particularly in terms of model efficiency and the substantial cost associated with manual annotation. In this work, we introduce a novel framework that integrates active learning within hyperbolic space to effectively address these issues. By leveraging the hierarchical representational capacity of hyperbolic space, which is naturally suited for modeling structured data, and combining it with the selective efficiency of active learning, our method introduces hyperbolic uncertainty metrics to guide the targeted selection of the most informative video frames and sequences for annotation. This enables the model to prioritize annotation efforts where they are most impactful. Furthermore, the model iteratively refines pseudo labels using all available annotations, significantly reducing the need for exhaustive labeling while preserving high segmentation accuracy. To further mitigate reliance on precise annotations, we enhance the MS-TCN model by incorporating soft pseudo labels and a weighting mechanism that dynamically adjusts learning based on label confidence, allowing for more robust training in the presence of noisy or weakly labeled data. Extensive experiments conducted on two widely used action segmentation benchmark datasets validate the effectiveness of our approach, demonstrating that it can substantially reduce annotation effort while maintaining overall segmentation performance.
- 24 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (7) , 1-18
Open AccessHyperbolic Active Learning for Label-Efficient Action Segmentation
Recent advances in action segmentation have greatly enhanced our understanding of complex and dynamic scenes in video content. Despite these improvements, the field continues to face persistent challenges, particularly in terms of model efficiency and the substantial cost associated with manual annotation. In this work, we introduce a novel framework that integrates active learning within hyperbolic space to effectively address these issues. By leveraging the hierarchical representational capacity of hyperbolic space, which is naturally suited for modeling structured data, and combining it with the selective efficiency of active learning, our method introduces hyperbolic uncertainty metrics to guide the targeted selection of the most informative video frames and sequences for annotation. This enables the model to prioritize annotation efforts where they are most impactful. Furthermore, the model iteratively refines pseudo labels using all available annotations, significantly reducing the need for exhaustive labeling while preserving high segmentation accuracy. To further mitigate reliance on precise annotations, we enhance the MS-TCN model by incorporating soft pseudo labels and a weighting mechanism that dynamically adjusts learning based on label confidence, allowing for more robust training in the presence of noisy or weakly labeled data. Extensive experiments conducted on two widely used action segmentation benchmark datasets validate the effectiveness of our approach, demonstrating that it can substantially reduce annotation effort while maintaining overall segmentation performance.
- Open Access
- 26 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 21-29
Mamba-Based Multimodal Continual Learning for Audio-Visual Classification with Prototype-Enhanced Anti-Forgetting Mechanism
- 26 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 21-29
Open AccessMamba-Based Multimodal Continual Learning for Audio-Visual Classification with Prototype-Enhanced Anti-Forgetting Mechanism
- 5 November 2024
- Published by Elsevier in Computer Vision and Image Understanding
- Vol. 250
Leaf cultivar identification via prototype-enhanced learning
- 5 November 2024
- Published by Elsevier in Computer Vision and Image Understanding
- Vol. 250
Leaf cultivar identification via prototype-enhanced learning
- Open Access
- 29 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 49-57
Dynamic Network for Language-based Fashion Retrieval
Language-based fashion image retrieval, as a kind of composed image retrieval, presents a substantial challenge in the domain of multi-modal retrieval. This task aims to retrieve the target fashion item in the gallery given a reference image and a modification text. Existing approaches primarily concentrate on developing a static multi-modal fusion module to learn the combined semantics of the reference image and modification text. Despite their commendable advancements, these approaches are still limited by a deficiency in flexibility, which is attributed to the application of a singluar fusion module across diverse input queries. In contrast to static fusion methods, we propose a novel method termed Dynamic Fusion Network (DFN) to compose the multi-granularity features dynamically by considering the consistency of routing path and modality-specific information simultaneously. In specific, our proposed method is consisted of two modules: (1) Dynamic Network. The dynamic network enables a flexible combination of the different operation modules, providing multi-granularity modality interaction for each reference image and modifier text. (2) Modality Specific Routers (MSR). The MSR generates precise routing decisions based on the distinct semantics and distributions of each reference image and modifier text. Extensive experiments on three benchmarks,\textiti.e., FashionIQ, Shoes and Fashion200K, demonstrate the effectiveness of our proposed model compared with the existing methods.
- 29 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 49-57
Open AccessDynamic Network for Language-based Fashion Retrieval
Language-based fashion image retrieval, as a kind of composed image retrieval, presents a substantial challenge in the domain of multi-modal retrieval. This task aims to retrieve the target fashion item in the gallery given a reference image and a modification text. Existing approaches primarily concentrate on developing a static multi-modal fusion module to learn the combined semantics of the reference image and modification text. Despite their commendable advancements, these approaches are still limited by a deficiency in flexibility, which is attributed to the application of a singluar fusion module across diverse input queries. In contrast to static fusion methods, we propose a novel method termed Dynamic Fusion Network (DFN) to compose the multi-granularity features dynamically by considering the consistency of routing path and modality-specific information simultaneously. In specific, our proposed method is consisted of two modules: (1) Dynamic Network. The dynamic network enables a flexible combination of the different operation modules, providing multi-granularity modality interaction for each reference image and modifier text. (2) Modality Specific Routers (MSR). The MSR generates precise routing decisions based on the distinct semantics and distributions of each reference image and modifier text. Extensive experiments on three benchmarks,\textiti.e., FashionIQ, Shoes and Fashion200K, demonstrate the effectiveness of our proposed model compared with the existing methods.
- Open Access
CFNet: Learning Correlation Functions for One-Stage Panoptic Segmentation
Recently, there is growing attention on one-stage panoptic segmentation methods which aim to segment instances and stuff jointly within a fully convolutional pipeline efficiently. However, most of the existing works directly feed the backbone features to various segmentation heads ignoring the demands for semantic and instance segmentation are different: The former needs semantic-level discriminative features, while the latter requires features to be distinguishable across instances. To alleviate this, we propose to first predict semantic-level and instance-level correlations among different locations that are utilized to enhance the backbone features, and then feed the improved discriminative features into the corresponding segmentation heads, respectively. Specifically, we organize the correlations between a given location and all locations as a continuous sequence and predict it as a whole. Considering that such a sequence can be extremely complicated, we adopt Discrete Fourier Transform (DFT), a tool that can approximate an arbitrary sequence parameterized by amplitudes and phrases. For different tasks, we generate these parameters from the backbone features in a fully convolutional way which is optimized implicitly by corresponding tasks. As a result, these accurate and consistent correlations contribute to producing plausible discriminative features which meet the requirements of the complicated panoptic segmentation task. To verify the effectiveness of our methods, we conduct experiments on several challenging panoptic segmentation datasets and achieve state-of-the-art performance on MS COCO with $45.1$\% PQ and ADE20k with $32.6$\% PQ.
Open AccessCFNet: Learning Correlation Functions for One-Stage Panoptic Segmentation
Recently, there is growing attention on one-stage panoptic segmentation methods which aim to segment instances and stuff jointly within a fully convolutional pipeline efficiently. However, most of the existing works directly feed the backbone features to various segmentation heads ignoring the demands for semantic and instance segmentation are different: The former needs semantic-level discriminative features, while the latter requires features to be distinguishable across instances. To alleviate this, we propose to first predict semantic-level and instance-level correlations among different locations that are utilized to enhance the backbone features, and then feed the improved discriminative features into the corresponding segmentation heads, respectively. Specifically, we organize the correlations between a given location and all locations as a continuous sequence and predict it as a whole. Considering that such a sequence can be extremely complicated, we adopt Discrete Fourier Transform (DFT), a tool that can approximate an arbitrary sequence parameterized by amplitudes and phrases. For different tasks, we generate these parameters from the backbone features in a fully convolutional way which is optimized implicitly by corresponding tasks. As a result, these accurate and consistent correlations contribute to producing plausible discriminative features which meet the requirements of the complicated panoptic segmentation task. To verify the effectiveness of our methods, we conduct experiments on several challenging panoptic segmentation datasets and achieve state-of-the-art performance on MS COCO with $45.1$\% PQ and ADE20k with $32.6$\% PQ.
- Open Access
TextRay: Contour-based Geometric Modeling for Arbitrary-shaped Scene Text Detection
Arbitrary-shaped text detection is a challenging task due to the complex geometric layouts of texts such as large aspect ratios, various scales, random rotations and curve shapes. Most state-of-the-art methods solve this problem from bottom-up perspectives, seeking to model a text instance of complex geometric layouts with simple local units (e.g., local boxes or pixels) and generate detections with heuristic post-processings. In this work, we propose an arbitrary-shaped text detection method, namely TextRay, which conducts top-down contour-based geometric modeling and geometric parameter learning within a single-shot anchor-free framework. The geometric modeling is carried out under polar system with a bidirectional mapping scheme between shape space and parameter space, encoding complex geometric layouts into unified representations. For effective learning of the representations, we design a central-weighted training strategy and a content loss which builds propagation paths between geometric encodings and visual content. TextRay outputs simple polygon detections at one pass with only one NMS post-processing. Experiments on several benchmark datasets demonstrate the effectiveness of the proposed approach. The code is available at https://github.com/LianaWang/TextRay.
Open AccessTextRay: Contour-based Geometric Modeling for Arbitrary-shaped Scene Text Detection
Arbitrary-shaped text detection is a challenging task due to the complex geometric layouts of texts such as large aspect ratios, various scales, random rotations and curve shapes. Most state-of-the-art methods solve this problem from bottom-up perspectives, seeking to model a text instance of complex geometric layouts with simple local units (e.g., local boxes or pixels) and generate detections with heuristic post-processings. In this work, we propose an arbitrary-shaped text detection method, namely TextRay, which conducts top-down contour-based geometric modeling and geometric parameter learning within a single-shot anchor-free framework. The geometric modeling is carried out under polar system with a bidirectional mapping scheme between shape space and parameter space, encoding complex geometric layouts into unified representations. For effective learning of the representations, we design a central-weighted training strategy and a content loss which builds propagation paths between geometric encodings and visual content. TextRay outputs simple polygon detections at one pass with only one NMS post-processing. Experiments on several benchmark datasets demonstrate the effectiveness of the proposed approach. The code is available at https://github.com/LianaWang/TextRay.
- 25 September 2014
- Published by Springer Nature in Neural Computing and Applications
- Vol. 26 (1) , 161-170
Location selection using fuzzy-connective-based aggregation networks: a case study of the food and beverage chain industry in Taiwan
- 25 September 2014
- Published by Springer Nature in Neural Computing and Applications
- Vol. 26 (1) , 161-170
Location selection using fuzzy-connective-based aggregation networks: a case study of the food and beverage chain industry in Taiwan