A5100431014- Neural Networks and Applications
- Rock Mechanics and Modeling
- Hydrocarbon exploration and reservoir analysis
- Soil, Finite Element Methods
- Grouting, Rheology, and Soil Mechanics
- Soil and Unsaturated Flow
- Seismic Imaging and Inversion Techniques
- EEG and Brain-Computer Interfaces
- Machine Learning in Healthcare
- Face and Expression Recognition
- Vehicle License Plate Recognition
- Ophthalmology and Visual Impairment Studies
- Domain Adaptation and Few-Shot Learning
- Image Processing Techniques and Applications
- Rough Sets and Fuzzy Logic
- Advanced Image Fusion Techniques
- Data Mining Algorithms and Applications
- BIM and Construction Integration
- Infrared Target Detection Methodologies
- Tactile and Sensory Interactions
- Multimodal Machine Learning Applications
- Remote-Sensing Image Classification
- Imbalanced Data Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Human Pose and Action Recognition
Institute of Art
2024
Xi'an Jiaotong University
2024
Hong Kong University of Science and Technology
2024
University of Hong Kong
2024
Changchun University of Science and Technology
2023
China University of Geosciences
2021
Okayama University
2019
Xiamen University
2017
Fujian University of Technology
2014
China National Petroleum Corporation (China)
2014
Functional Magnetic Resonance Imaging (fMRI) presents challenges due to limited paired samples and low signal-to-noise ratios, particularly in tasks involving reconstructing natural images or decoding their semantic content. To address these challenges, we introduce BrainCLIP, an innovative fMRI-based brain model. BrainCLIP leverages Contrastive Language-Image Pre-training's (CLIP) cross-modal generalization abilities bridge activity, images, text for the first time. Our experiments...
Due to the lack of paired samples and low signal-to-noise ratio functional MRI (fMRI) signals, reconstructing perceived natural images or decoding their semantic contents from fMRI data are challenging tasks. In this work, we propose, for first time, a task-agnostic fMRI-based brain model, BrainCLIP, which leverages CLIP's cross-modal generalization ability bridge modality gap between activity, image, text. Our experiments demonstrate that CLIP can act as pivot generic tasks, including...
Deciphering visual content from functional Magnetic Resonance Imaging (fMRI) helps illuminate the human vision system. However, scarcity of fMRI data and noise hamper brain decoding model performance. Previous approaches primarily employ subject-specific models, sensitive to training sample size. In this paper, we explore a straightforward but overlooked solution address scarcity. We propose shallow adapters map cross-subject into unified representations. Subsequently, shared deeper decodes...
Patients with amyotrophic lateral sclerosis ( ALS ) often have difficulty in expressing their intentions through language and behavior, which prevents them from communicating properly the outside world seriously affects quality of life. The brain-computer interface (BCI) has received much attention as an aid for patients to communicate world, but heavy device causes inconvenience application process. To improve portability BCI system, this paper proposed a wearable P300-speller system based...
Study on BIM Family Self-create for Steel Reinforcing Bar Detail Construction Design and Information Extraction Li-Chuan Lien, Pan-Cheng Zhang, S.B Chen, Z.C. Liao Y.N. Liu Pages 430-436 (2017 Proceedings of the 34rd ISARC, Taipei, Taiwan, ISBN 978-80-263-1371-7, ISSN 2413-5844) Abstract: In this study, frequently encountered issues related to reinforced concrete nodes (beam girder protection layers, beam column binding wire collision, etc.) were sorted out analyzed. Building Modeling (BIM)...
In this paper, the Deep Residual Network (ResNet) with Dempster-Shafer (D-S) evidence theory is presented for bimodal emotion recognition through applying facial expression and speech information. By acquiring discriminative features performing fusion of emotions, method can overcome limitations single modal obtain higher accuracy. The key areas emotional spectrograms are firstly used to acquire low-level characteristics emotion. Moreover, two ResNets designed select high-level semantic...
Mining Rules for Satellite Imagery Using Evolutionary Classification Tree L.C. Lien, Y.N. Liu, M.Y. Cheng, I-C. Yeh Pages 689-696 (2014 Proceedings of the 31st ISARC, Sydney, Australia, ISBN 978-0-646-59711-9, ISSN 2413-5844) Abstract: (CT) can establish explicit classification rules (SI). However, accuracy are poor. Back-Propagation Networks (BPN) and Support Vector Machine (SVM) both a highly accurate model to predict SI but cannot generate rules. This study proposes novel mining rule...
Welcome to the WIT Press eLibrary - home of Transactions Wessex Institute collection, providing on-line access papers presented at Institute's prestigious international conferences and from its State-of-the-Art in Science & Engineering publications.
Macular disease patients who lost central vision must use their peripheral to read slowly. There are some assistive methods proposed improve reading experience. Sequence word display like RSVP suited the performance of low people because it can reduce crowding effect, limit visual span and do not need saccades. However, these in is satisfactory. This study aimed design a assistant system that uses new method by fixed gaze patient test effects software called NRSVP produced improvement...