Interlaced Sparse Self-Attention for Semantic Segmentation
This addresses efficiency bottlenecks for semantic segmentation practitioners, though it appears incremental as it optimizes an existing mechanism rather than introducing a new paradigm.
The paper tackles the computational inefficiency of self-attention in semantic segmentation by proposing an interlaced sparse self-attention method that factorizes the dense affinity matrix into two sparse matrices, significantly reducing computation and memory complexity. The approach was empirically validated on six challenging semantic segmentation benchmarks.
In this paper, we present a so-called interlaced sparse self-attention approach to improve the efficiency of the \emph{self-attention} mechanism for semantic segmentation. The main idea is that we factorize the dense affinity matrix as the product of two sparse affinity matrices. There are two successive attention modules each estimating a sparse affinity matrix. The first attention module is used to estimate the affinities within a subset of positions that have long spatial interval distances and the second attention module is used to estimate the affinities within a subset of positions that have short spatial interval distances. These two attention modules are designed so that each position is able to receive the information from all the other positions. In contrast to the original self-attention module, our approach decreases the computation and memory complexity substantially especially when processing high-resolution feature maps. We empirically verify the effectiveness of our approach on six challenging semantic segmentation benchmarks.