提出多轴低秩视觉LSTM，提升医学图像分割精度与效率
MaLViL: Multi-axis Low-rank Vision-LSTM for Medical Image Segmentation
- 设计多轴低秩结构，在解码器多分辨率保持全局建模
- 在皮肤病变、超声和多器官CT上实现领先分割效果
- 内存占用减少83倍，适合高分辨率医学图像分析
视觉LSTM（ViL）虽能高效建模全局信息，但其计算成本随空间标记数量增长，现有分割模型将其限制在粗粒度瓶颈，导致细节丢失。将2D特征展开为1D序列会破坏正交扫描方向上的邻接关系。本文提出MaLViL，一种多轴低秩视觉LSTM网络，将ViL扩展至解码器多分辨率。双向低秩ViL（Bi-LRViL）在紧凑正交子空间中推理，并通过正交残差保留细节；尺度感知SaLViL在序列化前恢复跨轴邻居；交叉方向混合器（CDM）融合水平与垂直遍历路径。统计引导跳跃调制（SGSM）进一步保留编码器跳跃中的边界线索。在皮肤病变、超声及多器官CT基准测试中，MaLViL达到竞争性或最先进的分割精度，同时在精细解码器分辨率下使ViL算子内存降低高达83倍。代码已开源：https://github.com/xmindflow/malvil。
原文摘要 · Abstract (English)
Vision-LSTM (ViL) enables efficient global modeling, but its cost still scales with the number of spatial tokens, so existing segmenters confine ViL to a coarse bottleneck and lose fine anatomical detail. Rasterizing 2D features into a 1D sequence further breaks adjacency across the orthogonal scan axis. We propose MaLViL, a Multi-axis Low-rank Vision-LSTM network that extends ViL across decoder resolutions. Bidirectional low-rank ViL (Bi-LRViL) reasons on a compact orthonormal subspace and preserves detail through an orthogonal residual; scale-aware SaLViL restores cross-axis neighbors before serialization; and a Cross-Directional Mixer (CDM) fuses orthogonal horizontal and vertical traversal paths. Statistics-Guided Skip Modulation (SGSM) further retains boundary cues in encoder skips. On skin-lesion, ultrasound, and multi-organ CT benchmarks, MaLViL achieves competitive or state-of-the-art segmentation accuracy, while reducing ViL operator memory by up to $83\times$ at fine decoder resolutions. Code is available at: https://github.com/xmindflow/malvil.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。