果实沙拉:一套与风格一致的艺术作品数据集,以揭示图像嵌入中的相似感知
Tillmann Ohm1,2, Andres Karjus3,4,5, Mikhail V Tamm6,4
1Tallinn University, School of Digital Technologies, Tallinn, Estonia. mail@tillmannohm.com.
Scientific data
|February 12, 2025
概括
一个新的数据集,风格对齐的艺术品数据集 (SALAD),解决了缺乏视觉相似性的基准. 萨拉德能够对计算模型如何感知图像中的语义和风格相似性进行量化分析.
科学领域:
- 计算机视觉 计算机视觉
- 机器学习 机器学习
- 人工智能的人工智能
背景情况:
- 视觉相似性对于计算机视觉和图像嵌入模型至关重要.
- 现有的基准数据集很少,这阻碍了研究模型对相似性的感知.
研究的目的:
- 介绍风格对齐的艺术品数据集 (SALAD) 框架.
- 在计算模型中提供一个平衡的基准来评估语义和风格相似感知.
主要方法:
- 开发了Fruit-SALAD,这是一个数据集,包含10个类别和10种风格的10,000张水果图像.
- 利用生成图像合成的系统管道来创建数据集.
- 评估了各种计算模型,包括机器学习和特征提取算法.
主要成果:
- 在经过测试的模型中,SALAD在语义类别和风格相似性权重方面表现出显著的差异.
- 基准指标有助于对类似感知进行可量化和可解释的比较.
结论:
- 萨拉德提供了一个控制和平衡的平台,用于对相似感知进行比较分析.
- 允许对模型如何处理语义和风格相似性的强大,可量化和可解释的见解.
相关概念视频
01:58Fruit Development, Structure, and Function
22.0K
Fruits form from a mature flower ovary. As seeds develop from the ovules contained within, the ovary wall undergoes a series of complex changes to form fruit. In some fruits, such as soybeans, the ovary wall dries; in other fruits, such as grapes, it remains fleshy. In some cases, organs other than the ovary contribute to fruit formation; such fruits are called accessory fruits.
22.0K
02:13The Representativeness Heuristic
15.8K
The representative heuristic describes a biased way of thinking, in which you unintentionally stereotype someone or something. For example, you may assume that your professors spend their free time reading books and engaging in intellectual conversation, because the idea of them spending their time playing volleyball or visiting an amusement park does not fit in with your stereotypes of professors.
15.8K
02:16Light Acquisition
8.4K
In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.
8.4K
01:14Relative Frequency Histogram
5.4K
The relative frequency depicts the proportion of data points that have each value. The frequency tells the number of data points that have each value. Like the histogram, a relative frequency histogram also has the same shape with a horizontal scale (the x-axis), but the vertical scale (the y-axis) is marked with relative frequencies (percentages of the whole) instead of actual frequencies. A relative frequency histogram is a graphical representation of a frequency distribution where the...
5.4K
01:04Pie Chart
13.7K
A pie chart (or a pie graph) is a circular graphical chart or a pictorial representation of categorical data. It is divided into slices of pie each indicating numerical proportions. It is also used to show the relative sizes of data in a single chart.
In a pie chart, the central angle, the arc length of each slice, and the area are directly proportional to the quantity or percentage it represents. Some real-world examples that can be depicted using pie charts include marks obtained by students...
In a pie chart, the central angle, the arc length of each slice, and the area are directly proportional to the quantity or percentage it represents. Some real-world examples that can be depicted using pie charts include marks obtained by students...
13.7K