分析街景图像在美观性、活力度、安全性、无聊程度、富裕程度、压抑程度等维度的主观感知分数 Analyze subjective perception scores of street view images across dimensions like Beautiful, lively, safe, boring, wealthy, and depressing.
ConvNeXt+Segformer-B5+RF · Place Pulse 2.0 Dataset
分析公园社交媒体文本在可达性、可用性、吸引力、安全感、美感与维护管理等维度的感知分数 Analyze the perception scores of park-related social media texts across dimensions including accessibility, usability, attractiveness, safety, aesthetic, and maintenance.
Chinese-RoBERTa-wwm-ext · Park Perception Dataset先进的深度学习技术，精准的感知分析能力 Advanced deep learning technology with precise perception analysis capabilities
基于大规模数据集训练的深度学习模型，提供高精度的感知分析结果 Deep learning models trained on large-scale datasets provide high-precision perception analysis results
优化的模型推理引擎，实现秒级分析响应，提升用户体验 Optimized model inference engine achieves second-level analysis response, enhancing user experience
支持多个感知维度的独立分析，满足不同研究需求 Supports independent analysis of multiple perception dimensions to meet diverse research needs