MPL：面向少样本动作识别的匹配引导原型学习
MPL: Match-guided Prototype Learning for Few-shot Action Recognition
把匹配结果反向注入原型学习，并用关键帧级匹配细化原型，为少样本视频任务提供了一种「匹配-原型」联合优化的可复用范式；其跨样本注意力聚合器也可迁移到其他需要类内共享模式聚合的小样本视频设定。
原文依据 we propose a Match-guided Prototype Learning (MPL) method comprising two key components: enhanced match (E-Match) and key-frame extraction match (K-Match)
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论当前少样本动作识别方法通过学习具有代表性的原型并设计多样化的视频匹配策略取得了令人瞩目的性能。然而，这些方法通常面临两个关键局限：i）通过隐式样本交互学习到的原型在查询-支持对之间缺乏清晰的语义对应关系，限制了其类别代表性；ii）原型学习与匹配机制的独立设计导致原型表示与匹配策略之间存在潜在的不兼容性。为解决上述局限，我们提出匹配引导原型学习（MPL）方法，包含两个关键组件：增强匹配（E-Match）与关键帧提取匹配（K-Match）。E-Match 通过引入查询样本的匹配语义，在类别特定嵌入中显式增强原型学习；K-Match 则在细粒度帧级别上通过关键帧匹配进一步细化原型表示。此外，我们提出跨样本注意力聚合器（CSA-Aggregator），动态聚合支持样本之间的相邻帧，从而获得能够捕捉类内共享动作模式的原型表示。由此，所提出的 MPL 能够有效地从查询-支持对中挖掘由粗到细、匹配引导的语义信息，以生成具有判别性的类别原型，并提升原型表示与匹配机制的兼容性。在四个公开数据集上的大量评估证实，MPL 优于领先的少样本动作识别技术。源代码见 https://github.com/jayzh-research/MPL-FSAR。
英文原文摘要 1711 字符
Current few-shot action recognition methods achieve impressive performance by learning representative prototypes and designing diverse video matching strategies. However, these approaches typically face two critical limitations: i) prototypes learned through implicit sample interactions lack clear semantic correspondence between query-support pairs, limiting their class representativeness; ii) the independent design of prototype learning and matching mechanisms creates a potential incompatibility between prototype representations and matching strategies. To address these limitations, we propose a Match-guided Prototype Learning (MPL) method comprising two key components: enhanced match (E-Match) and key-frame extraction match (K-Match). E-Match explicitly enhances prototype learning in class-specific embeddings by incorporating the matched semantics of query samples, while K-Match further refines the prototype representation through key-frame matching at the fine-grained frame level. Additionally, we propose a Cross-Shot Attention Aggregator (CSA-Aggregator) that dynamically aggregates adjacent frames across support samples, thereby obtaining a prototype representation that captures intra-class shared action patterns. In this way, the proposed MPL effectively mines coarse-to-fine, match-guided semantic information from query-support pairs to generate discriminative class prototypes, and improve the compatibility of prototype representation with the match mechanism. Extensive evaluations on four public datasets confirm that MPL achieves superior performance over leading few-shot action recognition techniques. The source code is available at https://github.com/jayzh-research/MPL-FSAR.