中国古代“瓶”类器物视觉美学与文本知识向量集数据

1. 特定器型演变研究：为研究“瓶”这一器型在不同朝代、不同窑口的造型、尺寸、纹饰、工艺的演变规律，提供量化分析的数据基础。 2. AI辅助器型识别与分类：训练能够自动识别图片中器物是否为“瓶”，并能进一步区分其子类型（如梅瓶、胆瓶等）的AI模型。 3. 三维重建与数字孪生应用：高精度的视觉向量可作为AI进行3D形态重建的先验知识，辅助生成“瓶”类器物的数字孪生模型，用于虚拟展览或修复研究。1. Research on the Evolution of Specific Vessel Types: This dataset provides a quantitative analysis data foundation for studying the evolution patterns of the shape, size, decoration, and craftsmanship of the "bottle" vessel type across different dynasties and kiln sites. 2. AI-assisted Vessel Recognition and Classification: It is used to train AI models that can automatically identify whether an artifact in an image is a "bottle", and further distinguish its subtypes such as plum vase (meiping), gallbladder vase (danping), etc. 3. 3D Reconstruction and Digital Twin Applications: High-precision visual vectors can serve as prior knowledge for AI to conduct 3D shape reconstruction, assisting in generating digital twin models of "bottle" artifacts for virtual exhibitions or restoration research.

搜集汇总

背景与挑战

背景概述

该数据集是中国古代'瓶'类器物的专题数据集，包含多模态特征向量（4048维），来源于拍卖行内部数据，通过器物学分类和关键词匹配提取。它支持器型演变研究、AI识别分类和三维重建等应用，数据格式为CSV，处理规则包括形态识别和向量生成。