Camera Style Adaptation for Person Re-identification

Being a cross-camera retrieval task, person re-identification suffers from image style variations caused by different cameras. The art implicitly addresses this problem by learning a camera-invariant descriptor subspace... (read more)

PDF Abstract CVPR 2018 PDF CVPR 2018 Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK BENCHMARK
Person Re-Identification DukeMTMC-reID IDE* Rank-1 72.31 # 54
MAP 51.83 # 56
Person Re-Identification DukeMTMC-reID IDE* + CamStyle + Random Erasing Rank-1 78.32 # 48
MAP 57.61 # 51
Person Re-Identification Market-1501 IDE* Rank-1 85.66 # 62
MAP 65.87 # 67
Person Re-Identification Market-1501 IDE* + CamStyle + Random Erasing Rank-1 89.49 # 54
MAP 71.55 # 60

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