面向目标检测的保持驱动知识压缩黑盒域自适应
Black-Box Domain Adaptation for Object Detection with Retention-Driven Knowledge Compression
把终身学习中的记忆保持与场景压缩机制引入黑盒域自适应目标检测，用分区记忆对比学习过滤噪声伪标签、用近远视图对比挖掘远视区域线索，可直接迁移到无源数据的目标检测域迁移场景。
原文依据 Black-Box Domain Adaptation (BBDA) is a highly practical yet challenging strategy that enables the deployment of pre-trained detectors to new unlabeled target domains
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论黑盒域自适应（BBDA）是一种高度实用但颇具挑战性的策略，它使预训练检测器能够在无法访问源数据或源模型的情况下部署到新的无标注目标域。与以往的域自适应研究相比，BBDA 不仅提供了更强的数据隐私保护，还具有更好的可移植性。尽管相关研究日益增多，现有 BBDA 策略仍难以直接应用于目标检测，因为此前大多数工作集中于不涉及边界框定位的分类与分割任务，且依赖不同的学习机制。本文受终身学习启发，提出保持驱动知识压缩（RDKC），将类脑持续学习过程应用于面向目标检测的 BBDA。具体而言，RDKC 由两个关键组件构成：记忆保持（MR）与场景压缩（SC）。MR 专门针对 BBDA 设定下的目标检测设计，它在划分出的区域上执行记忆化对比学习，从而更好地利用可靠区域中的信息性线索，同时过滤掉噪声预测标签带来的潜在噪声。SC 在近视图与远视图区域之间引入对比机制，使模型能够在近视图线索的引导下更好地从远视图区域学习。实验结果表明，在 BBDA 设定下，RDKC 在所有评测基准上均优于此前的 SOTA 方法，取得了更优的性能提升。
英文原文摘要 1523 字符
Black-Box Domain Adaptation (BBDA) is a highly practical yet challenging strategy that enables the deployment of pre-trained detectors to new unlabeled target domains without accessing source data or models. Compared to previous domain adaptation studies, BBDA not only provides stronger data privacy protection but also offers greater portability. Despite growing interest, existing BBDA strategies remain difficult to apply directly to object detection, as most prior works focus on classification and segmentation tasks that do not involve bounding box localization and rely on different learning mechanisms. In this paper, inspired by lifelong learning, we propose Retention-Driven Knowledge Compression (RDKC), which applies a brain-inspired continual learning process to BBDA for object detection. Specifically, RDKC consists of two key components: Memory Retention (MR) and Scene Compression (SC). MR is designed specifically for object detection under the BBDA setting, where it performs memorized contrastive learning on partitioned regions to better utilize informative cues from reliable areas while filtering out potential noise from noise prediction labels. SC introduces a contrastive mechanism between near- and far-view regions, which enables the model to better learn from far-view regions under the guidance of near-view cues. Experimental results demonstrate that under the BBDA setting, RDKC outperforms previous SOTA methods across all evaluated benchmarks, achieving superior performance improvements.