Fine-grained visual categorization via multi-stage metric learning

By Qi Qian, Rong Jin, Shenghuo Zhu, Yuanqing Lin. 2014

155 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Representation & Self-Supervised Learning

Citations in the corpus

Growing a Brain: Fine-Tuning by Increasing Model Capacity (2017, neutral): “CNN 62.9 Finetuning-CNN 57.7 Caffe [53] 59.5 CNN-SVM [32] 74.7 CNN-SVM [32] 53.3 Deep Standard [4] 58.9 — — CNNaug-SVM [32] 86.8 CNNaug-SVM [32] 61.8 — — Task Customized CNNs Caffe-DAG [53] 64.6 LSVM [30] 87.1 LSVM [30] 61.4 Deep Optimized [4] 66.4 — — MsML+ [30] 89.5 DeCaf+DPD [7] 65.0 — — Places-C”
Deep Metric Learning via Lifted Structured Feature Embedding (2015, background): “This function poses two computational challenges: (1) it is non-smooth, and (2) both evaluating it and computing the subgradient requires mining all pairs of examples several times.”
Metric Learning with Adaptive Density Discrimination (2015, methodology): “Beyond classiﬁcation, we further validate the saliency of the learnt representations via their attribute concentration and hierarchy recovery properties, achieving 10-25% relative gains on the softmax classiﬁer and 25-50% on triplet loss in these tasks.”
Fine-Grained Image Classification by Exploring Bipartite-Graph Labels (2015, background): “[44] showed a more discriminative similarity can be learned through deep ranking and metric learning respectively.”
Object-Part Attention Model for Fine-Grained Image Classification (2017, background): “…92.51 VGGNet NAC [20] 91.60 VGGNet FOAF [8] 91.39 VGGNet ONE+SVM [50] 90.03 VGGNet Deep Optimized [51] 88.10 AlexNet NAC [20] 85.20 AlexNet MsML+ [52] 81.18 CNN MsML [52] 80.45 CNN Deep Standard [51] 78.50 AlexNet Shape+Appearance [3] 56.68 Zernike+SCC [53] 59.50 GMP+p [54] 56.80 GMP [54] 56.10…”
Fine-Grained Categorization and Dataset Bootstrapping Using Deep Metric Learning with Humans in the Loop (2015, background): “This section presents a bootstrapping framework on how to grow a small scale, fine-grained dataset in an efficient manner.”
Integrating Scene Text and Visual Appearance for Fine-Grained Image Classification (2017, background): “A multi-stage distance metric learning framework is proposed in [23] to learn a more discriminative similarity.”
RepMet: Representative-based metric learning for classification and one-shot object detection (2018, methodology): “We tested our approach on a set of fine-grained classification datasets, as is customary for the state-of-the-art DML classification works [24, 39, 21]: Stanford Dogs [14], Oxford-IIIT Pet [20], Oxford 102 Flowers [19], and ImageNet Attributes [24].”
Transfer of View-manifold Learning to Similarity Perception of Novel Objects (2017, methodology): “DCNNs, such as AlexNet, pre-trained on large dataset, have developed useful feature representations that can be fine-tuned for other specific tasks (Donahue et al., 2014; Qian et al., 2015; Karpathy et al., 2014).”
Cross-Dimensional Weighting for Aggregated Deep Convolutional Features (2015, background): “The state-of-the-art on fine-grained classification has advanced rapidly during the last few years, with many approaches either training on or mining object parts [30, 17, 27].”
Fully Convolutional Attention Localization Networks: Efficient Attention Localization for Fine-Grained Recognition (2016, background): “Qian et al. [19] 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.”
Adaptive Triplet Model for Fine-Grained Visual Categorization (2018, methodology): “It is interesting to compare our model with two triplet-related models: DMLHIL [27] and MSML [36].”
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.”
Learning Deep Embeddings with Histogram Loss (2016, background): “Triplet-based losses are popular for large-scale embedding learning [4] and in particular for deep embeddings [13, 14, 17, 21, 28].”
SoftTriple Loss: Deep Metric Learning Without Triplet Sampling (2019, background): “Later, various strategies are developed to reduce the computational cost [16, 17].”
RepMet: Representative-Based Metric Learning for Classification and Few-Shot Object Detection (2018, methodology): “We also demonstrate the use of our approach for general DML-based classification by comparing to the Magnet Loss [25] and other state-of-the-art DML-based approaches [43, 22].”
DarkRank: Accelerating Deep Metric Learning via Cross Sample Similarities Transfer (2017, background): “In recent years, end-to-end deep metric learning method which learns feature representation by the guide of metric based losses has achieved great success(Qian et al. 2015; Song et al. 2016; Schroff, Kalenichenko, and Philbin 2015).”