Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identification

Person re-identification (re-ID) models trained on one domain often fail to generalize well to another. In our attempt, we present a "learning via translation" framework... (read more)

PDF Abstract CVPR 2018 PDF CVPR 2018 Abstract
No code implementations yet. Submit your code now

Datasets


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Person Re-Identification DukeMTMC-reID SPGAN+LMP* Rank-1 46.4 # 50
MAP 26.2 # 57
Unsupervised Person Re-Identification DukeMTMC-reID SPGAN+LMP Rank-1 46.4 # 3
Rank-10 68.0 # 3
Rank-5 62.3 # 3
MAP 26.2 # 3
Unsupervised Domain Adaptation Duke to Market SPGAN mAP 22.8 # 20
rank-1 51.5 # 20
rank-5 70.1 # 13
rank-10 76.8 # 15
Unsupervised Person Re-Identification Market-1501 SPGAN+LMP Rank-1 57.7 # 4
MAP 26.7 # 3
Rank-10 82.4 # 3
Rank-5 75.8 # 3
Unsupervised Domain Adaptation Market to Duke SPGAN mAP 22.3 # 19
rank-1 41.1 # 17
rank-5 56.6 # 14
rank-10 63.0 # 13
Unsupervised Person Re-Identification MSMT17->DukeMTMC-reID SPGAN Rank-1 46.4 # 2
Rank-10 68.0 # 2
Rank-5 62.3 # 2
mAP 26.2 # 2

Methods used in the Paper


METHOD TYPE
Batch Normalization
Normalization
Residual Connection
Skip Connections
PatchGAN
Discriminators
ReLU
Activation Functions
Tanh Activation
Activation Functions
Residual Block
Skip Connection Blocks
Instance Normalization
Normalization
Convolution
Convolutions
Leaky ReLU
Activation Functions
Sigmoid Activation
Activation Functions
GAN Least Squares Loss
Loss Functions
Cycle Consistency Loss
Loss Functions
CycleGAN
Generative Models
Siamese Network
Twin Networks