• Corpus ID: 1396647

In Defense of the Triplet Loss for Person Re-Identification

  title={In Defense of the Triplet Loss for Person Re-Identification},
  author={Alexander Hermans and Lucas Beyer and B. Leibe},
In the past few years, the field of computer vision has gone through a revolution fueled mainly by the advent of large datasets and the adoption of deep convolutional neural networks for end-to-end learning. [] Key Result We show that, for models trained from scratch as well as pretrained ones, using a variant of the triplet loss to perform end-to-end deep metric learning outperforms most other published methods by a large margin.

Person Re-id by Incorporating PCA Loss in CNN

This paper proposes an algorithm, particularly a loss function and its end to end learning manner, for person re-identification task. The main idea is to take full advantage of the labels in a batch

Efficient Feature Extraction for Person Re-Identification via Distillation

This work proposes using deep neural network distillation for training a feature extractor with a lower computational cost, while keeping track of its cross-domain ability, which is three times faster, without a decrease in accuracy.

Incremental Learning in Person Re-Identification

A model that can be used for multiple tasks in Person Re-Identification, provide state-of-the-art results on a variety of tasks and still achieve considerable accuracy subsequently is proposed.

Discriminative Feature Representation for Person Re-identification by Batch-contrastive Loss

This work introduces a new auxiliary loss function, called batch-contrastive loss, for person reID to further separate the features of different identities and pulls the Features of same identity closer.

Autoencoder Ensemble for Person Re-Identification

A simple yet effective autoencoder, comprising of an encoder and a sequential decoder, that can make features more generalizable to the unknown test data to prevent from overfitting.

Improving Deep Models of Person Re-identification for Cross-Dataset Usage

This work develops a method of training the model on multiple datasets, and a method for its online practically unsupervised fine-tuning, which yield up to 19.1% improvement in Rank-1 score in the cross-dataset evaluation.

Person re-identification using visual attention

A novel approach based on using a gradient-based attention mechanism in deep convolution neural network for solving the person re-identification problem by learns to focus selectively on parts of the input image for which the networks' output is most sensitive to.

Exhaustive hard triplet mining loss for Person Re-Identification

This work proposes a novel variant of the triplet loss, named exhaustive hard triplet mining loss (EHTM), which is able to deal with various forms of hard triplets in a comprehensive manner and provides an effective training strategy to further enhance model performance.

Bags of Tricks and A Strong Baseline for Deep Person Re-identification.

A simple and efficient baseline for person re-identification (ReID) with deep neural networks is explored, which achieves 94.5% rank-1 and 85.9% mAP on Market1501 with only using global features.

Learning Incremental Triplet Margin for Person Re-identification

The margin between positive and negative pairs of triplets is explored and it is proved that large margin is beneficial and a novel multi-stage training strategy which learns incremental triplet margin and improves triplet loss effectively is proposed.



Multi-Scale Triplet CNN for Person Re-Identification

A multi-scale triplet convolutional neural network which captures visual appearance of a person at various scales is proposed, addressing the problem of small training set in person re-identification.

Gated Siamese Convolutional Neural Network Architecture for Human Re-identification

A gating function is proposed to selectively emphasize such fine common local patterns that may be essential to distinguish positive pairs from hard negative pairs by comparing the mid-level features across pairs of images.

Person Re-identification by Multi-Channel Parts-Based CNN with Improved Triplet Loss Function

A novel multi-channel parts-based convolutional neural network model under the triplet framework for person re-identification that significantly outperforms many state-of-the-art approaches, including both traditional and deep network-based ones, on the challenging i-LIDS, VIPeR, PRID2011 and CUHK01 datasets.

Joint Learning for Attribute-Consistent Person Re-Identification

This work proposes an approach that goes beyond appearance by integrating a semantic aspect into the model, and outperforms several state-of-the-art methods on VIPeR, a standard re-identification dataset.

A Discriminatively Learned CNN Embedding for Person Reidentification

This article proposes a Siamese network that simultaneously computes the identification loss and verification loss and learns a discriminative embedding and a similarity measurement at the same time, thus taking full usage of the re-ID annotations.

Person Re-identification: Past, Present and Future

The history of person re-identification and its relationship with image classification and instance retrieval is introduced and two new re-ID tasks which are much closer to real-world applications are described and discussed.

Joint Learning of Single-Image and Cross-Image Representations for Person Re-identification

This work proposes a joint learning frame-work to unify SIR and CIR using convolutional neural network (CNN), and finds that the representations learned with pairwise comparison and triplet comparison objectives can be combined to improve matching performance.

Deep Attributes Driven Multi-Camera Person Re-identification

By directly using the deep attributes with simple Cosine distance, this work has obtained surprisingly good accuracy on four person ReID datasets, and shows that a simple metric learning modular further boosts the method, making it significantly outperform many recent works.