Open access
Jul 2026
PAST: Prior-Aware Sparse Transformer for Micro-Expression Recognition
The Prior-aware Sparse Transformer (PAST) is proposed, a novel Transformer-based architecture integrating spatial and semantic prior knowledge synergistically into a sparse attention mechanism, enabling linear-complexity processing of large amounts of fine-grained features.
Jiateng Liu, Tianchen Zhou, Heng-Can Shi et al.
· Electronics · 0 citations