SETransformer combines a Transformer encoder, channel attention, and attention pooling for WISDM activity recognition, but the architecture is permutation-invariant and the reported comparison omits the model itself.
An overview of indoor localization system for human activity recognition (har) in healthcare
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
REJECT 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
SETransformer: A Hybrid Attention-Based Architecture for Robust Human Activity Recognition
SETransformer combines a Transformer encoder, channel attention, and attention pooling for WISDM activity recognition, but the architecture is permutation-invariant and the reported comparison omits the model itself.