Unsupervised Template Learning for Fine-Grained Object Recognition

By Shulin Yang, Liefeng Bo, Jue Wang, Linda G. Shapiro. 2012

167 citations recorded by the source corpus.

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

Citations in the corpus

Fine-Grained Image Classification via Combining Vision and Language (2017, neutral): “Object annotation is used in the works of [12, 13] to learn part detectors in an unsupervised or latent manner.”
Part-Based R-CNNs for Fine-Grained Category Detection (2014, methodology): “Previous work has investigated part-based approaches to this problem [7, 16, 30, 44, 47, 22].”
The application of two-level attention models in deep convolutional neural network for fine-grained image classification (2014, methodology): “learn part detectors in unsupervised or latent manner [23, 5].”
The Unreasonable Effectiveness of Noisy Data for Fine-Grained Recognition (2015, background): “The majority of research in fine-grained recognition has focused on developing improved models for classification [1,3,5,7,9,8,14,16,18,20,21,22,28,29,36,37,41,42,49,51,50,66,68,69,71,73,72,76,77,75,78].”
Neural Activation Constellations: Unsupervised Part Model Discovery with Convolutional Networks (2015, background): “[37] select object part templates from a set of randomly initialized image patches.”
Bird Species Categorization Using Pose Normalized Deep Convolutional Nets (2014, background): “Areas explored include feature representations that better preserve fine-grained information [35, 46, 47, 48], segmentation-based approaches [1, 13, 14, 15, 21, 37] that facilitate extraction of purer features, and part/pose normalized feature spaces [5, 6, 19, 33, 38, 39, 43, 50, 51].”
Attention for Fine-Grained Categorization (2014, neutral): “ortly. 3 Under review as a workshop contribution at ICLR 2015 Table 1: Previous state-of-the-art results on Stanford Dogs test set, measured by mean accuracy (mA) as described in Chai et al. (2013)2. Yang et al. (2012) 0.38 Chai et al. (2013) 0.46 Gavves et al. (2013) 0.50 GoogLeNet 96 96 0.42 GoogL”
Symbiotic Segmentation and Part Localization for Fine-Grained Categorization (2013, background): “It successfully pulls together a number of recent research strands: the use of distinctive parts for registration when discriminating sub-ordinate categories [5, 24, 32, 34, 35]; unsupervised discovery of midlevel discriminative patches [23, 28, 32]; learning a DPM given only weak annotation (a loos”
Augmenting Strong Supervision Using Web Data for Fine-Grained Categorization (2015, methodology): “Various methods have been used to define object parts such as unsupervised patch discovery [32, 8], human-in-the-loop methods [12, 10], or direct reliance on strongly supervised datasets with part-level annotations [21, 4, 33, 20].”
Local Alignments for Fine-Grained Categorization (2014, methodology): “Following common practice in the ﬁne-grained literature (Yao et al. 2011, 2012; Yang et al. 2012) we mirror the training images in the datasets to double the size of the training set.”
Picking Deep Filter Responses for Fine-Grained Image Recognition (2016, background): “Match [29] bbox bbox 38% Symbolic [3] bbox bbox 45.”
Learning Features and Parts for Fine-Grained Recognition (2014, background): “[5], [22], [23]) or mining discriminative templates [5], [6].”
Fine-grained visual categorization via multi-stage metric learning (2014, methodology): “The feature extraction step, which sometimes combines with segmentation [1, 7, 24], part localization [2, 35] or both [6], is to extract image level representations, and popular choices include LLC features [1], Fisher vectors [14], etc.”
Fully Convolutional Attention Networks for Fine-Grained Recognition (2016, background): “[26] propose a multi-stage metric learning framework to learn a distance metric that pulls data points of the same class close and pushes data points from different classes far apart.”
Hierarchical Part Matching for Fine-Grained Visual Categorization (2013, background): “To cope with the fine-grained classification tasks, researchers have proposed many novel algorithms, such as visual attributes [8], random templates [22], hierarchical matching [6] and part-based one-vs-one features [2].”
Fine-grained recognition without part annotations (2015, background): “Though many early approaches [15, 49, 50, 48] did not use part annotations, their performance has been eclipsed by methods developed to explicitly take advantage of the structure present in fine-grained classes [3, 6, 51, 53, 13].”
Coarse-to-Fine Description for Fine-Grained Visual Categorization (2016, result): “Similar with DPM, Yang et al. [33] propose an unsupervised template matching strategy to detect certain parts.”
Integrating Scene Text and Visual Appearance for Fine-Grained Image Classification (2017, background): “Examples are the recent works in [10], [11] that use salient parts in images.”