Modeling Image Patches with a Generic Dictionary of Mini-epitomes
By George Papandreou, Liang-Chieh Chen, Alan L. Yuille. 2014
19 citations recorded by the source corpus.
The goal of this paper is to question the necessity of features like SIFT in categorical visual recognition tasks. As an alternative, we develop a generative model for the raw intensity of image patches and show that it can support image classification performance on par with optimized SIFT-based techniques in a bag-of-visual-words setting. Key ingredient of the proposed model is a compact dictionary of mini-epitomes, learned in an unsupervised fashion on a large collection of images. The use of epitomes allows us to explicitly account for photometric and position variability in image appearance. We show that this flexibility considerably increases the capacity of the dictionary to accurately approximate the appearance of image patches and support recognition tasks. For image classification, we develop histogram-based image encoding methods tailored to the epitomic representation, as well as an "epitomic footprint" encoding which is easy to visualize and highlights the generative nature of our model. We discuss in detail computational aspects and develop efficient algorithms to make the model scalable to large tasks. The proposed techniques are evaluated with experiments on the challenging PASCAL VOC 2007 image classification benchmark.
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
- Alan L. Yuille
- Luc Van Gool
- Dacheng Tao
- Trevor Darrell
- Andrea Vedaldi
- Xiaogang Wang
- Li Fei-Fei
- Jitendra Malik
Citations in the corpus
- ImageNet Large Scale Visual Recognition Challenge (2014, background): “3 - - Toyota Technological Institute at Chicago†, Ecole Centrale Paris‡ George Papandreou†, Iasonas Kokkinos‡(Papandreou, 2014; Papandreou et al., 2014; Jojic et al., 2003; Krizhevsky et al., 2012; Sermanet et al., 2013; Dubout and Fleuret, 2012; Iandola et al., 2014)”
- Deep Epitomic Convolutional Neural Networks (2014, methodology): “Recently, single-level mini-epitomes learned by a variant of K-means have been prop osed as an alternative to SIFT for image classification [23].”
- Visual Concepts and Compositional Voting (2017, methodology): “Attempts have been made to obtain building blocks by studying properties of image patches [30] and [5]”
- Modeling local and global deformations in Deep Learning: Epitomic convolution, Multiple Instance Learning, and sliding window detection (2015, methodology): “Recently, singlelevel mini-epitomes learned by a variant of K-means have been proposed as an alternative to SIFT for image classification [35].”
- Inverse Graphics with Probabilistic CAD Models (2014, methodology): “Image comparison could be improved by using richer appearance models such as Epitomes [12].”