Hard-Aware Deeply Cascaded Embedding

@article{Yuan2017HardAwareDC,
  title={Hard-Aware Deeply Cascaded Embedding},
  author={Yuhui Yuan and Kuiyuan Yang and Chao Zhang},
  journal={2017 IEEE International Conference on Computer Vision (ICCV)},
  year={2017},
  pages={814-823}
}
  • Yuhui Yuan, Kuiyuan Yang, Chao Zhang
  • Published 17 November 2016
  • Computer Science
  • 2017 IEEE International Conference on Computer Vision (ICCV)
Riding on the waves of deep neural networks, deep metric learning has achieved promising results in various tasks by using triplet network or Siamese network. Though the basic goal of making images from the same category closer than the ones from different categories is intuitive, it is hard to optimize the objective directly due to the quadratic or cubic sample size. Hard example mining is widely used to solve the problem, which spends the expensive computation on a subset of samples that are… 
View With Semantic Reader
220 Citations
Highly Influential Citations
39
Background Citations
105
Methods Citations
77
Results Citations
13

Figures and Tables from this paper

220 Citations

Citation Type

Citation Type

LoOp: Looking for Optimal Hard Negative Embeddings for Deep Metric Learning
TLDR
This work proposes a novel approach that looks for optimal hard negatives (LoOp) in the embedding space, taking full advantage of each tuple by calculating the minimum distance between a pair of positives and a Pair of negatives.
Assignment Problem Based Deep Embedding
TLDR
This paper proposes a novel linear assignment problem based hard sample mining strategy for contrastive loss to learn feature embeddings and can obtain the state-of-the-art performance on the CUB-200-2011, Cars196, and In-shop datasets with the GoogLeNet network.
An Adversarial Approach to Hard Triplet Generation
TLDR
This work proposes an adversarial network for Hard Triplet Generation (HTG) to optimize the network ability in distinguishing similar examples of different categories as well as grouping varied examples of the same categories.
The Group Loss for Deep Metric Learning
TLDR
Group Loss is proposed, a loss function based on a differentiable label-propagation method that enforces embedding similarity across all samples of a group while promoting, at the same time, low-density regions amongst data points belonging to different groups.
Deep Metric Learning by Online Soft Mining and Class-Aware Attention
TLDR
This work proposes a novel sample mining method, called Online Soft Mining (OSM), which assigns one continuous score to each sample to make use of all samples in the mini-batch, and introduces Class-Aware Attention (CAA) that assigns little attention to abnormal data samples.
SoftTriple Loss: Deep Metric Learning Without Triplet Sampling
TLDR
The SoftTriple loss is proposed to extend the SoftMax loss with multiple centers for each class, equivalent to a smoothed triplet loss where each class has a single center.
Learning Intra-Batch Connections for Deep Metric Learning
TLDR
This work proposes an approach based on message passing networks that takes into account all the relations in a mini-batch of samples into account, and refine embedding vectors by exchanging messages among all samples in a given batch allowing the training process to be aware of the overall structure.
Smart Mining for Deep Metric Learning
TLDR
This paper proposes a novel deep metric learning method that combines the triplet model and the global structure of the embedding space and relies on a smart mining procedure that produces effective training samples for a low computational cost.
Hard Example Mining with Auxiliary Embeddings
TLDR
The experiments on the challenging Disguised Faces in the Wild dataset show that hard example mining with auxiliary embeddings improves the discriminative power of learned representations.
The Group Loss++: A deeper look into group loss for deep metric learning
TLDR
Group Loss is proposed, a loss function based on a differentiable label-propagation method that enforces embedding similarity across all samples of a group while promoting, at the same time, low-density regions amongst data points belonging to different groups.
...
...

References

SHOWING 1-10 OF 53 REFERENCES
Deep Metric Learning via Lifted Structured Feature Embedding
TLDR
An algorithm for taking full advantage of the training batches in the neural network training by lifting the vector of pairwise distances within the batch to the matrix of Pairwise distances enables the algorithm to learn the state of the art feature embedding by optimizing a novel structured prediction objective on the lifted problem.
Improved Deep Metric Learning with Multi-class N-pair Loss Objective
TLDR
This paper proposes a new metric learning objective called multi-class N-pair loss, which generalizes triplet loss by allowing joint comparison among more than one negative examples and reduces the computational burden of evaluating deep embedding vectors via an efficient batch construction strategy using only N pairs of examples.
Embedding Label Structures for Fine-Grained Feature Representation
TLDR
The proposed multitask learning framework significantly outperforms previous fine-grained feature representations for image retrieval at different levels of relevance and to model the multi-level relevance, label structures such as hierarchy or shared attributes are seamlessly embedded into the framework by generalizing the triplet loss.
Deeply-Supervised Nets
TLDR
The proposed deeply-supervised nets (DSN) method simultaneously minimizes classification error while making the learning process of hidden layers direct and transparent, and extends techniques from stochastic gradient methods to analyze the algorithm.
BranchyNet: Fast inference via early exiting from deep neural networks
TLDR
The BranchyNet architecture is presented, a novel deep network architecture that is augmented with additional side branch classifiers that can both improve accuracy and significantly reduce the inference time of the network.
Zero-Shot Learning via Joint Latent Similarity Embedding
TLDR
A joint discriminative learning framework based on dictionary learning is developed to jointly learn the parameters of the model for both domains, which ultimately leads to a class-independent classifier that shows 4.90% improvement over the state-of-the-art in accuracy averaged across four benchmark datasets.
Fine-Grained Categorization and Dataset Bootstrapping Using Deep Metric Learning with Humans in the Loop
TLDR
Experimental evaluations show significant performance gain using dataset bootstrapping and demonstrate state-of-the-art results achieved by the proposed deep metric learning methods.
Learning Deep Embeddings with Histogram Loss
TLDR
It is shown that these operations can be performed in a simple and piecewise-differentiable manner using 1D histograms with soft assignment operations, which makes the proposed loss suitable for learning deep embeddings using stochastic optimization.
Learning Local Image Descriptors with Deep Siamese and Triplet Convolutional Networks by Minimizing Global Loss Functions
TLDR
A combination of the triplet and global losses produces the best embedding in the field, using this triplet network, and it is demonstrated that the use of the central-surround siamese network trained with the global loss producing the best result of the field on the UBC dataset.
Local Similarity-Aware Deep Feature Embedding
TLDR
This paper introduces a Position-Dependent Deep Metric (PDDM) unit, which is capable of learning a similarity metric adaptive to local feature structure that can be used to select genuinely hard samples in a local neighborhood to guide the deep embedding learning in an online and robust manner.
...
...