A mixed high-order attention network built from factorized polynomial predictors improves person re-identification accuracy on three benchmarks over prior state-of-the-art methods.
Higher-order in- tegration of hierarchical convolutional activations for fine- grained visual categorization
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Mixed High-Order Attention Network for Person Re-Identification
A mixed high-order attention network built from factorized polynomial predictors improves person re-identification accuracy on three benchmarks over prior state-of-the-art methods.