Attributes as Operators

نویسندگان

  • Tushar Nagarajan
  • Kristen Grauman
چکیده

We present a new approach to modeling visual attributes. Prior work casts attributes in a similar role as objects, learning a latent representation where properties (e.g., sliced) are recognized by classifiers much in the way objects (e.g., apple) are. However, this common approach fails to separate the attributes observed during training from the objects with which they are composed, making it ineffectual when encountering new attribute-object compositions. Instead, we propose to model attributes as operators. Our approach learns a semantic embedding that explicitly factors out attributes from their accompanying objects, and also benefits from novel regularizers expressing attribute operators’ effects (e.g., blunt should undo the effects of sharp). Not only does our approach align conceptually with the linguistic role of attributes as modifiers, but it also generalizes to recognize unseen compositions of objects and attributes. We validate our approach on two challenging datasets and demonstrate significant improvements over the state-of-the-art. In addition, we show that not only can our model recognize unseen compositions robustly in an open-world setting, it can also generalize to compositions where objects themselves were unseen during training.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Trapezoidal intuitionistic fuzzy prioritized aggregation operators and application to multi-attribute decision making

In some multi-attribute decision making (MADM) problems, various relationships among the decision attributes should be considered. This paper investigates the prioritization relationship of attributes in MADM with trapezoidal intuitionistic fuzzy numbers (TrIFNs). TrIFNs are a special intuitionistic fuzzy set on a real number set and have the better capability to model ill-known quantities. Fir...

متن کامل

Port Performance Analysis Using Extent Fuzzy AHP Approach

Many methods have been adopted by shipping operators to evaluate, analyze and select ports of calls. Majority of these methods are generally based on evaluating different port performance criteria involved. The ports performance criteria are interrelated and are of a complex nature. Comparison of ports by different evaluation methods provides a basis for implementing operating strategies for...

متن کامل

Attribute Based Perceptual Mapping of Prepaid Mobile Cellular Operators: An Empirical Investigation Among Management Graduates in Central Haryana

A comparison of consumer’s perception regarding cellular mobile operators was done on five cellular operators in the prepaid segment. The paper focuses on how management students associate selected brands of mobile cellular operators. Discriminant analysis is used to map selected mobile cellular operators along with their attributes. Study investigates how consumers associate the attributes in ...

متن کامل

Generalized Interval-Valued Atanassov’s Intuitionistic Fuzzy Power Operators and Their Application to Group Decision Making

We generalize the power averaging operators to interval-valued Atanassov’s intuitionistic fuzzy environments, and develop a series of generalized interval-valued Atanassov’s intuitionistic fuzzy power aggregation operators. The main advantages of these operators are that they not only accommodate situations in which the input arguments are interval-valued intuitionistic fuzzy numbers (IVIFNs), ...

متن کامل

Intelligent Knowledge Based System Approach for Optimization of Design and Manufacturing for Abrasive Water Jet Machining

A water jet machining is an industrial tool capable of cutting a wide variety of materials using a very high-pressure jet of water, or a mixture of water and an abrasive substance. This paper addresses the concept of the Intelligent knowledge base system (IKBS) for optimization of product design and manufacturing process for water jet machining in computer based concurrent engineering environme...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2018