Weakly Supervised Fine-Grained Image Categorization

By Yu Zhang, Xiu-Shen Wei, Jianxin Wu, Jianfei Cai, Jiangbo Lu, Viet Anh Nguyen, Minh N. Do. 2015

53 citations recorded by the source corpus.

In this paper, we categorize fine-grained images without using any object / part annotation neither in the training nor in the testing stage, a step towards making it suitable for deployments. Fine-grained image categorization aims to classify objects with subtle distinctions. Most existing works heavily rely on object / part detectors to build the correspondence between object parts by using object or object part annotations inside training images. The need for expensive object annotations prevents the wide usage of these methods. Instead, we propose to select useful parts from multi-scale part proposals in objects, and use them to compute a global image representation for categorization. This is specially designed for the annotation-free fine-grained categorization task, because useful parts have shown to play an important role in existing annotation-dependent works but accurate part detectors can be hardly acquired. With the proposed image representation, we can further detect and visualize the key (most discriminative) parts in objects of different classes. In the experiment, the proposed annotation-free method achieves better accuracy than that of state-of-the-art annotation-free and most existing annotation-dependent methods on two challenging datasets, which shows that it is not always necessary to use accurate object / part annotations in fine-grained image categorization.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

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 [2, 5, 7, 15, 17, 18, 24, 30, 37, 44, 45, 58, 59, 60, 63, 64].”
Part-Stacked CNN for Fine-Grained Visual Categorization (2015, unknown): “Constellation [36] n/a n/a 68.5 Attention [48] n/a n/a 69.7 Bilinear-CNN [22] n/a n/a 74.2 Weak FGVC [54] n/a n/a 75.0 CNNaug [31] BBox BBox 61.8 Alignment [13] BBox BBox 67.0 No parts [18] BBox BBox 74.9 Bilinear-CNN [22] BBox BBox 80.4 Part R-CNN [51] BBox+Parts n/a 73.9 PoseNorm CNN [6]…”
Adaptive Triplet Model for Fine-Grained Visual Categorization (2018, background): “Therefore, two-step models [5]–[7], [13]–[19] believe that the key to FGVC is finding distinct areas with discriminative information.”
Fully Convolutional Attention Localization Networks: Efficient Attention Localization for Fine-Grained Recognition (2016, methodology): “Zhang et al. [25] select useful parts from multi-scale part proposals in objects, and use them to compute a global image representation for categorization.”
Visual Concept Recognition and Localization via Iterative Introspection (2016, unknown): “92[24] 77 .”
A review on the main challenges in automatic plant disease identification based on visible range images (2016, unknown): “Also, some techniques allow the problem to be formulated directly as an image categorization task, in which context deep convolutional neural networks have played an important role (Lin, RoyChowdhury, & Maji, 2015; Zhang, Donahue, Girshick, & Darrell, 2014; Zhang et al., 2015).”
Audio Visual Attribute Discovery for Fine-Grained Object Recognition (2018, background): “This is better than several baselines (Zhang et al. 2015; Lin, RoyChowdhury, and Maji 2015).”