A5048225135- Domain Adaptation and Few-Shot Learning
- Advanced Image and Video Retrieval Techniques
- Advanced Neural Network Applications
- Sustainable Building Design and Assessment
- Regulation and Compliance Studies
- Visual Attention and Saliency Detection
- Multimodal Machine Learning Applications
- Environmental Sustainability in Business
- COVID-19 diagnosis using AI
China University of Mining and Technology
2023-2024
YOLOX is a state-of-the-art one-stage object detection model for real-time applications that employs decoupled head and advanced label assignment. Despite its impressive performance, has limitations prevent it from achieving optimal accuracy in settings. To improve these limitations, we propose new approach called re-parameterization align (RA-YOLOX). Our novel to the classification regression tasks, enhancing learning of connection information between regression. In addition, assignment(LA)...
The visual Transformer model based on self-attention has achieved better performance than convolutional neural networks in object detection tasks. However, existing models are typically heavy-weight to extract global features. In contrast, CNNs can features with fewer parameters and computational costs. To combine the advantages of processing at local level Transformer's interaction, this paper proposes MCANet, a Hierarchical Cross-Fusion Lightweight Based Multi-ConvHead Attention for Object...
Cross-modal remote sensing image-text retrieval (CMRSITR) aims to extract comprehensive information from diverse modalities. The primary challenge in this field is developing effective mappings between visual and textual modalities a shared latent space. Existing approaches generally focus on utilizing pre-trained unimodal models independently features each modality. However, these techniques often fall short achieving the critical alignment necessary for cross-modal matching. These...
The divergence of environmental, social, and governance (ESG) ratings across providers is an area increasingly greater focus given their increased use by regulators for policymaking investors investment decisions. Here, the authors discuss at a granular level what goes into ESG rating, particularly modeling choices involved in construction. provide step-by-step illustration using companies global automobiles industry with “raw” data from four leading providers. They differences metrics...