面向红外小目标检测的重建引导目标感知可逆编码器
Target-Aware Invertible Encoder with Reconstruction Guidance for Infrared Small Target Detection
可逆编码器将信息损失显式化为可优化量，配合仅在逆路径生效的目标感知重建调制，为小目标检测提供了一种不改变前向检测分布的表征增强手段，可迁移到其他尺度敏感检测任务。
原文依据 An invertible pathway reconstructs the input from feature latents, exposing information loss as an optimizable quantity.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论现代检测器通常加深骨干网络并依赖激进的下采样来获取高层语义。但这会通过重缩放引起的信息丢失严重退化低能量红外小目标。本文提出InvDet，一种目标感知可逆编码器，在重建引导的检测框架内统一了信息保留与目标感知增强。可逆通路从特征潜变量重建输入，将信息丢失暴露为可优化量。为了将检测与无关重建解耦，目标感知重建调制（TARM）模块仅在逆路径中操作，对高通潜变量进行门控并对低通特征施加温和增益，而不改变前向检测分布。此外，提出了几何-内容容忍度量（GCTM）以聚焦真正有信息的区域，并产生逐像素权重图来温和正则化重建分支。我们的方法在五个公开红外基准上取得了有竞争力的准确率，同时展现出强大的跨数据集泛化能力，为尺度受限的视觉场景提供了一条通往检测友好表征学习的原则性路径。
英文原文摘要 1221 字符
Modern detectors typically deepen backbones and rely on aggressive downsampling to harvest high-level semantics. But this severely degrades low-energy infrared tiny targets via rescale-induced information loss. This work introduces InvDet, a target-aware invertible encoder that unifies information preservation and target-aware enhancement within a reconstruction-guided detection framework. An invertible pathway reconstructs the input from feature latents, exposing information loss as an optimizable quantity. To decouple detection from irrelevant reconstruction, a Target-Aware Reconstruction Modulation (TARM) module operates only in the inverse path, gating high-pass latents and applying a mild gain to low-pass features without altering the forward detection distribution. In addition, a Geometry-Content Tolerance Metric (GCTM) is proposed to focus on truly informative regions and yields a pixel-wise weight map that gently regularizes the reconstruction branch. Our method achieves competitive accuracy on five public infrared benchmarks while exhibiting strong cross-dataset generalization, providing a principled pathway toward detection-friendly representation learning for scale-challenged visual regimes.