HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Vehicle Re Identification On Vehicleid Large

평가 지표

Rank-1

Rank-5

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	Rank-1	Rank-5	Paper Title	Repository
PNP Loss	93.2	96.6	Rethinking the Optimization of Average Precision: Only Penalizing Negative Instances before Positive Ones is Enough
Recall@k Surrogate loss (ViT-B/16)	94.7	97.1	Recall@k Surrogate Loss with Large Batches and Similarity Mixup
ANet	80.5	94.6	AttributeNet: Attribute Enhanced Vehicle Re-Identification	-
Smooth-AP	91.9	96.2	Smooth-AP: Smoothing the Path Towards Large-Scale Image Retrieval
QD-DLF	-	-	Vehicle Re-identification Using Quadruple Directional Deep Learning Features	-
CAL	75.1	-	Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identification
MSINet (2.3M w/o RK)	77.9	91.7	MSINet: Twins Contrastive Search of Multi-Scale Interaction for Object ReID
RPTM	92.9	96.3	Relation Preserving Triplet Mining for Stabilising the Triplet Loss in Re-identification Systems
Recall@k Surrogate loss (ResNet-50)	93.8	96.6	Recall@k Surrogate Loss with Large Batches and Similarity Mixup
vehiclenet	79.46	-	VehicleNet: Learning Robust Feature Representation for Vehicle Re-identification

0 of 10 row(s) selected.