SOTAVerified

Metric Learning

The goal of Metric Learning is to learn a representation function that maps objects into an embedded space. The distance in the embedded space should preserve the objects’ similarity — similar objects get close and dissimilar objects get far away. Various loss functions have been developed for Metric Learning. For example, the contrastive loss guides the objects from the same class to be mapped to the same point and those from different classes to be mapped to different points whose distances are larger than a margin. Triplet loss is also popular, which requires the distance between the anchor sample and the positive sample to be smaller than the distance between the anchor sample and the negative sample.

Source: Road Network Metric Learning for Estimated Time of Arrival

Papers

Showing 51100 of 1648 papers

TitleStatusHype
Understanding the Role of the Projector in Knowledge DistillationCode1
End-to-end One-shot Human ParsingCode1
A Comparison of Metric Learning Loss Functions for End-To-End Speaker VerificationCode1
ePillID Dataset: A Low-Shot Fine-Grained Benchmark for Pill IdentificationCode1
Exploring Complementary Strengths of Invariant and Equivariant Representations for Few-Shot LearningCode1
Adversarial Background-Aware Loss for Weakly-supervised Temporal Activity LocalizationCode1
Versatile User Identification in Extended Reality using Pretrained Similarity-LearningCode1
Facial Expression Recognition in the Wild via Deep Attentive Center LossCode1
DA-VPT: Semantic-Guided Visual Prompt Tuning for Vision TransformersCode1
Manifold Matching via Deep Metric Learning for Generative ModelingCode1
Few-Shot Object Detection with Fully Cross-TransformerCode1
Few-Shot Open-Set Recognition using Meta-LearningCode1
Deep Iterative and Adaptive Learning for Graph Neural NetworksCode1
Continual Learning of Object InstancesCode1
Combining Metric Learning and Attention Heads For Accurate and Efficient Multilabel Image ClassificationCode1
Contrastive Bayesian Analysis for Deep Metric LearningCode1
Circle Loss: A Unified Perspective of Pair Similarity OptimizationCode1
Bridging the Gap: Multi-Level Cross-Modality Joint Alignment for Visible-Infrared Person Re-IdentificationCode1
A Benchmark and Asymmetrical-Similarity Learning for Practical Image Copy DetectionCode1
Close Imitation of Expert Retouching for Black-and-White PhotographyCode1
Contrastive Learning with Hard Negative SamplesCode1
A Unified Object Motion and Affinity Model for Online Multi-Object TrackingCode1
Characterizing Generalization under Out-Of-Distribution Shifts in Deep Metric LearningCode1
AGKD-BML: Defense Against Adversarial Attack by Attention Guided Knowledge Distillation and Bi-directional Metric LearningCode1
Collaborative Translational Metric LearningCode1
Collapse-Aware Triplet Decoupling for Adversarially Robust Image RetrievalCode1
A Hybrid System of Sound Event Detection Transformer and Frame-wise Model for DCASE 2022 Task 4Code1
Compositional Learning of Image-Text Query for Image RetrievalCode1
Attributes-Guided and Pure-Visual Attention Alignment for Few-Shot RecognitionCode1
Cross-Batch Memory for Embedding LearningCode1
Cross Modal Retrieval with Querybank NormalisationCode1
Crystal structure prediction with machine learning-based element substitutionCode1
ABO: Dataset and Benchmarks for Real-World 3D Object UnderstandingCode1
DeCLUTR: Deep Contrastive Learning for Unsupervised Textual RepresentationsCode1
Deep Compositional Metric LearningCode1
A Metric Learning Reality CheckCode1
Deep Learning for Person Re-identification: A Survey and OutlookCode1
Deep Metric Learning by Online Soft Mining and Class-Aware AttentionCode1
Deep Metric Learning for Unsupervised Remote Sensing Change DetectionCode1
Deep metric learning using Triplet networkCode1
AbRank: A Benchmark Dataset and Metric-Learning Framework for Antibody-Antigen Affinity RankingCode1
Deformation-Aware 3D Model Embedding and RetrievalCode1
Back to the Feature: Learning Robust Camera Localization from Pixels to PoseCode1
Directional Statistics-based Deep Metric Learning for Image Classification and RetrievalCode1
An Inductive Bias for Distances: Neural Nets that Respect the Triangle InequalityCode1
Domain-invariant Similarity Activation Map Contrastive Learning for Retrieval-based Long-term Visual LocalizationCode1
3rd Place Solution to "Google Landmark Retrieval 2020"Code1
Dynamic Metric Learning: Towards a Scalable Metric Space to Accommodate Multiple Semantic ScalesCode1
A Deep Metric Learning Approach to Account LinkingCode1
Attribute-aware Identity-hard Triplet Loss for Video-based Person Re-identificationCode1
Show:102550
← PrevPage 2 of 33Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Unicom+ViT-L@336pxR@198.2Unverified
2Hyp-DINO 8x8R@192.8Unverified
3ResNet-50 + AVSLR@191.5Unverified
4NEDR@191.5Unverified
5ResNet-50 + Intra-Batch (ensemble of 5)R@191.5Unverified
6EfficientDML-VPTSP-G/512R@191.2Unverified
7CCL (ResNet-50)R@191.02Unverified
8ResNet50 + LanguageR@190.2Unverified
9ResNet-50 + MetrixR@189.6Unverified
10ResNet50 + S2SDR@189.5Unverified
#ModelMetricClaimedVerifiedStatus
1Unicom+ViT-L@336pxR@191.2Unverified
2STIRR@188.3Unverified
3Recall@k Surrogate Loss (ViT-B/16)R@188Unverified
4ViT-TripletR@186.5Unverified
5ROADMAP (DeiT-S)R@186Unverified
6Hyp-ViTR@185.9Unverified
7Hyp-DINOR@185.1Unverified
8Recall@k Surrogate Loss (ViT-B/32)R@185.1Unverified
9CCL (ResNet-50)R@183.1Unverified
10ROADMAP (ResNet-50)R@183.1Unverified
#ModelMetricClaimedVerifiedStatus
1Unicom+ViT-L@336pxR@190.1Unverified
2EfficientDML-VPTSP-G/512R@188.5Unverified
3Hyp-ViTR@185.6Unverified
4Hyp-DINOR@180.9Unverified
5NEDR@174.9Unverified
6CCL (ResNet-50)R@173.45Unverified
7ResNet-50 + AVSLR@171.9Unverified
8ResNet-50 + Intra-Batch ConnectionsR@171.8Unverified
9ResNet50 + LanguageR@171.4Unverified
10ResNet-50 + MetrixR@171.4Unverified
#ModelMetricClaimedVerifiedStatus
1Unicom+ViT-L@336pxR@196.7Unverified
2STIRR@195Unverified
3MGAR@194.3Unverified
4Hyp-ViTR@192.5Unverified
5Hyp-DINOR@192.4Unverified
6CCL (ResNet-50)R@192.31Unverified
7Gradient SurgeryR@192.21Unverified
8ResNet-50 + MetrixR@192.2Unverified
9EfficientDML-VPTSP-G/512R@192.1Unverified
10ViT-TripletR@192.1Unverified
#ModelMetricClaimedVerifiedStatus
1HAPPIERAverage-mAP43.8Unverified
2CSLAverage-mAP31Unverified
#ModelMetricClaimedVerifiedStatus
1HAPPIERAverage-mAP38Unverified
2CSLAverage-mAP28.7Unverified
#ModelMetricClaimedVerifiedStatus
1HAPPIERAverage-mAP37Unverified
2CSLAverage-mAP12.1Unverified