Measuring and Predicting Object Importance
By Merrielle Spain, Pietro Perona. 2011
83 citations recorded by the source corpus.
How important is a particular object in a photograph of a complex scene? We propose a definition of importance and present two methods for measuring object importance from human observers. Using this ground truth, we fit a function for predicting the importance of each object directly from a segmented image; our function combines a large number of object-related and image-related features. We validate our importance predictions on 2,841 objects and find that the most important objects may be identified automatically. We find that object position and size are particularly informative, while a popular measure of saliency is not.
Subfields: Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Trevor Darrell
- Xiaogang Wang
- Bernt Schiele
- Dacheng Tao
- Li Fei-Fei
- Alan L. Yuille
- Jitendra Malik
Citations in the corpus
- Semantic Understanding of Scenes Through the ADE20K Dataset (2016, background): “4 Dataset statistics Fig. 5.a shows the distribution of ranked object frequencies. The distribution is similar to a Zipf’s law and is typically found when objects are exhaustively annotated in images [32, 34]. They differ from the ones from datasets such as 6 Bolei Zhou et al. COCO or ImageNet where”
- Scene Parsing through ADE20K Dataset (2017, background): “The distribution is similar to a Zipf’s law and is typically found when objects are exhaustively annotated in images [29, 32].”
- Understanding and predicting importance in images (2012, background): “Perhaps most similar to our goals, elegant work by Spain and Perona [22] [23] tackles the task of examining factors to predict the order in which objects will be mentioned given an image.”
- VIP: Finding important people in images (2015, methodology): “We model importance in two ways: • Image-Level Importance: “Given an image, who is the most important individual?””
- Bringing Semantics into Focus Using Visual Abstraction (2013, methodology): “Nouns are the most commonly collected [29, 31] and studied part of speech.”
- Adopting Abstract Images for Semantic Scene Understanding (2016, methodology): “Many methods use tagged objects in images to predict important objects directly from visual features [3], [4], [5], [6], and to study the properties of popular tags [5], [6].”