HyperAI超神经

首页算力平台文档资讯论文教程数据集百科 SOTA LLM 模型天梯 GPU 天梯顶会

中文

HyperAI超神经

Vehicle Re Identification On Vehicleid Medium

评估指标

Rank-1

Rank-5

评测结果

各个模型在此基准测试上的表现结果

			Paper Title	Repository
Recall@k Surrogate loss (ViT-B/16)	95.2	97.2	Recall@k Surrogate Loss with Large Batches and Similarity Mixup
Recall@k Surrogate loss (ResNet-50)	94.6	96.9	Recall@k Surrogate Loss with Large Batches and Similarity Mixup
PNP Loss	94.2	96.9	Rethinking the Optimization of Average Precision: Only Penalizing Negative Instances before Positive Ones is Enough
Smooth-AP	93.3	96.4	Smooth-AP: Smoothing the Path Towards Large-Scale Image Retrieval
RPTM	93.3	96.5	Relation Preserving Triplet Mining for Stabilising the Triplet Loss in Re-identification Systems
ANet	82.8	96.2	AttributeNet: Attribute Enhanced Vehicle Re-Identification	-
vehiclenet	81.35	-	VehicleNet: Learning Robust Feature Representation for Vehicle Re-identification	-
CAL	78.2	-	Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identification
QD-DLF	-	-	Vehicle Re-identification Using Quadruple Directional Deep Learning Features	-

0 of 9 row(s) selected.

Vehicle Re Identification On Vehicleid Medium | SOTA | HyperAI超神经