Filters








2,961 Hits in 5.1 sec

Pairwise Confusion for Fine-Grained Visual Classification [article]

Abhimanyu Dubey, Otkrist Gupta, Pei Guo, Ramesh Raskar, Ryan Farrell, Nikhil Naik
2018 arXiv   pre-print
Fine-Grained Visual Classification (FGVC) datasets contain small sample sizes, along with significant intra-class variation and inter-class similarity.  ...  Our procedure, called Pairwise Confusion (PC) reduces overfitting by intentionally introducing confusion in the activations.  ...  Ashok Gupta for his guidance on bird recognition, and Dr. Sumeet Agarwal, Spandan Madan and Ishaan Grover for their feedback at various stages of this work.  ... 
arXiv:1705.08016v3 fatcat:juahdrhjgnci7f2ibq2k2ikrau

Pairwise Confusion for Fine-Grained Visual Classification [chapter]

Abhimanyu Dubey, Otkrist Gupta, Pei Guo, Ramesh Raskar, Ryan Farrell, Nikhil Naik
2018 Lecture Notes in Computer Science  
Fine-Grained Visual Classification (FGVC) datasets contain small sample sizes, along with significant intra-class variation and interclass similarity.  ...  This procedure, called Pairwise Confusion (PC) attempts to learn features with greater generalization, thereby preventing overfitting.  ...  To do so, we propose Pairwise Confusion (PC), a pairwise algorithm for training convolutional neural networks (CNNs) end-to-end for fine-grained visual classification.  ... 
doi:10.1007/978-3-030-01258-8_5 fatcat:lxujhpul5rexhi5nhhlknoeecy

Learning Attentive Pairwise Interaction for Fine-Grained Classification

Peiqin Zhuang, Yali Wang, Yu Qiao
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Fine-grained classification is a challenging problem, due to subtle differences among highly-confused categories.  ...  We conduct extensive experiments on five popular benchmarks in fine-grained classification.  ...  Hence, API-Net is a concise and practical deep learning approach for fine-grained classification. Figure 3 : Visualization.  ... 
doi:10.1609/aaai.v34i07.7016 fatcat:p6txgmvnjjhkrbos4b354cjkee

ACE: Adaptive Confusion Energy for Natural World Data Distribution [article]

Yen-Chi Hsu, Cheng-Yao Hong, Wan-Cyuan Fan, Ming-Sui Lee, Davi Geiger, Tyng-Luh Liu
2021 arXiv   pre-print
fine-grained and long-tailed properties at the same time.  ...  Most data in the natural world usually have imbalanced distribution and fine-grained characteristics.  ...  Introduction Fine-grained visual classification (FGVC) is an active and challenging problem in computer vision.  ... 
arXiv:1910.12423v3 fatcat:qgd7qrgp25bq5mlfwkw3pnkdgm

Learning Attentive Pairwise Interaction for Fine-Grained Classification [article]

Peiqin Zhuang, Yali Wang, Yu Qiao
2020 arXiv   pre-print
Fine-grained classification is a challenging problem, due to subtle differences among highly-confused categories.  ...  We conduct extensive experiments on five popular benchmarks in fine-grained classification.  ...  Attentive Pairwise Interaction In this section, we describe Attentive Pairwise Interaction Network (API-Net) for fine-grained classification.  ... 
arXiv:2002.10191v1 fatcat:ya57j4hb45arvdrdshv7y6gac4

Low-Rank Pairwise Alignment Bilinear Network For Few-Shot Fine-Grained Image Classification [article]

Huaxi Huang, Junjie Zhang, Jian Zhang, Jingsong Xu, Qiang Wu
2020 arXiv   pre-print
To filling the classification gap, in this paper, we address the Few-Shot Fine-Grained (FSFG) classification problem, which focuses on tackling the fine-grained classification under the challenging few-shot  ...  Nonetheless, it is still challenging for current FS models to distinguish the subtle differences between fine-grained categories given limited training data.  ...  Fine-Grained Object Classification Fine-grained object classification has been a trending topic in the computer vision research area for years, and most traditional fine-grained approaches use hand-crafted  ... 
arXiv:1908.01313v3 fatcat:ijrllxx4yfeu7oxtw3s752bvve

A new dataset of dog breed images and a benchmark for finegrained classification

Ding-Nan Zou, Song-Hai Zhang, Tai-Jiang Mu, Min Zhang
2020 Computational Visual Media  
In this paper, we introduce an image dataset for fine-grained classification of dog breeds: the Tsinghua Dogs Dataset.  ...  It is currently the largest dataset for fine-grained classification of dogs, including 130 dog breeds and 70,428 real-world images.  ...  Acknowledgements The authors would like to thank Wei-Yu Xie for his assistance on paper writing, and also thank Qiu Xin and Zhi-Ping Zhang for much help on image processing and labeling.  ... 
doi:10.1007/s41095-020-0184-6 fatcat:7xljdplr75gsvatpqgclxny22e

Hierarchical bilinear convolutional neural network for image classification

Xiang Zhang, Lei Tang, Hangzai Luo, Sheng Zhong, Ziyu Guan, Long Chen, Chao Zhao, Jinye Peng, Jianping Fan
2021 IET Computer Vision  
Image classification is one of the mainstream tasks of computer vision. However, the most existing methods use labels of the same granularity level for training.  ...  This leads to ignoring the hierarchy that may help to differentiate different visual objects better.  ...  features to fine-grained features, which may be helpful for image classification.  ... 
doi:10.1049/cvi2.12023 fatcat:apr3ebtwdreohivpdol6um72ai

Bilinear CNN Models for Fine-Grained Visual Recognition

Tsung-Yu Lin, Aruni RoyChowdhury, Subhransu Maji
2015 2015 IEEE International Conference on Computer Vision (ICCV)  
This architecture can model local pairwise feature interactions in a translationally invariant manner which is particularly useful for fine-grained categorization.  ...  We present experiments and visualizations that analyze the effects of fine-tuning and the choice two networks on the speed and accuracy of the models.  ...  on a popular fine-grained recognition dataset for birds [37] .  ... 
doi:10.1109/iccv.2015.170 dblp:conf/iccv/LinRM15 fatcat:6un34dm5u5d6ri2zsmt6ccu2xu

Relieving Long-tailed Instance Segmentation via Pairwise Class Balance [article]

Yin-Yin He, Peizhen Zhang, Xiu-Shen Wei, Xiangyu Zhang, Jian Sun
2022 arXiv   pre-print
In this paper, we explore to excavate the confusion matrix, which carries the fine-grained misclassification details, to relieve the pairwise biases, generalizing the coarse one.  ...  To this end, we propose a novel Pairwise Class Balance (PCB) method, built upon a confusion matrix which is updated during training to accumulate the ongoing prediction preferences.  ...  So we take one step further, a two-dimensional statistics (i.e., confusion matrix) is utilized to indicate fine-grained pairwise bias. Confusion Matrix.  ... 
arXiv:2201.02784v2 fatcat:gsdtba3xd5f7fhio2urgoyatua

Sentential negation of abstract and concrete conceptual categories: a brain decoding multivariate pattern analysis study

Marta Ghio, Karolin Haegert, Matilde M. Vaghi, Marco Tettamanti
2018 Philosophical Transactions of the Royal Society of London. Biological Sciences  
MVPA classification of the main effect of fine-grained conceptual category.  ...  (e) CP5: main effect of fine-grained conceptual category The mean whole-brain accuracy for the classification of the fine-grained conceptual categories was 52.78% (chance level: 100%/6 ¼ 16.67%) (figure  ...  We received no funding for this study.  ... 
doi:10.1098/rstb.2017.0124 pmid:29914992 fatcat:jvm35v2lo5hllhhqodzagk5sui

Learning Attention-Aware Interactive Features for Fine-Grained Vegetable and Fruit Classification

Yimin Wang, Zhifeng Xiao, Lingguo Meng
2021 Applied Sciences  
Vegetable and fruit recognition can be considered as a fine-grained visual categorization (FGVC) task, which is challenging due to the large intraclass variances and small interclass variances.  ...  Inspired by this intuition, a recent FGVC method, named Attentive Pairwise Interaction Network (API-Net), takes as input an image pair for pairwise feature interaction and demonstrates superior performance  ...  In contrast, humans often recognize fine-grained objects by comparing image pairs to extract subtle visual differences that can be used as distinguishable features.  ... 
doi:10.3390/app11146533 fatcat:i6qoyjgsvrh2lpuel2p4w4lfbu

Distribution Shift Metric Learning for Fine-grained Ship Classification in SAR Images

Haitao Lang, Yongjie Xu
2020 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Fine-grained ship classification in synthetic aperture radar (SAR) images is a challenging task, since SAR images can only provide limited discriminative information due to the limitation of SAR imaging  ...  Extensive experiments and in-depth analysis demonstrate that the proposed DML-ds can effectively increase the interclass separability and the intraclass compactness, thereby improving the fine-grained  ...  ACKNOWLEDGMENT The authors would like to thank all anonymous reviewers and associate editor for their constructive comments and suggestions that significantly improved this article.  ... 
doi:10.1109/jstars.2020.2991784 fatcat:4vww5w4ojzdpdpmwi3svror4y4

Predicting Violence Rating Based on Pairwise Comparison

Ying JI, Yu WANG, Jien KATO, Kensaku MORI
2020 IEICE transactions on information and systems  
Each video is annotated with 6 fine-grained objective attributes, which are considered to be closely related to violence extent.  ...  Experiment results on this dataset demonstrate the effectiveness of our method compared with the state-of-art classification methods.  ...  Additionally, each clip is manually labelled with 6 objective fine-grained visual attributes.  ... 
doi:10.1587/transinf.2020edp7056 fatcat:7jxl753s3jcrjkvt2iuduuixc4

Discovering localized attributes for fine-grained recognition

Kun Duan, D. Parikh, D. Crandall, K. Grauman
2012 2012 IEEE Conference on Computer Vision and Pattern Recognition  
They are particularly useful for fine-grained domains where categories are closely related to one other (e.g. bird species recognition). In such scenarios, relevant attributes are often local (e.g.  ...  In this paper, we propose an interactive approach that discovers local attributes that are both discriminative and semantically meaningful from image datasets annotated only with fine-grained category  ...  Acknowledgments: The authors thank Dhruv Batra for discussions on the L-CRF formulation, and acknowledge support from NSF IIS-1065390, the Luce Foundation, the Lilly Endowment, and the IU Data-to-Insight  ... 
doi:10.1109/cvpr.2012.6248089 dblp:conf/cvpr/DuanPCG12 fatcat:ih5iacw3cvb5fgd2n2cnt7p2ma
« Previous Showing results 1 — 15 out of 2,961 results