存储更少，查找更多：新颖性过滤如何改进边缘摄像头上的跨模态检索
Storing Less, Finding More: How Novelty Filtering Improves Cross-Modal Retrieval on Edge Cameras
AI总结：
本文提出一种流式检索架构，通过边缘设备的epsilon-net过滤器保留语义新颖帧，构建去噪嵌入索引，并结合跨模态适配器和云重排序器，提升边缘摄像头跨模态检索性能。
摘要：
始终开启的边缘摄像头生成连续视频流，冗余帧会通过拥挤将正确结果挤出top-k搜索。本文提出一种流式检索架构：边缘设备的epsilon-net过滤器仅保留语义新颖帧，构建去噪嵌入索引；跨模态适配器和云重排序器补偿紧凑编码器的弱对齐。在两个眼动数据集（AEA、EPIC-KITCHENS）上，单次流式过滤器在八个视觉-语言模型（8M-632M）中优于离线方法（k-means、最远点、均匀、随机）。结合后，该架构在8M边缘编码器下达到45.6%的Hold-out数据Hit@5，功耗估计为2.7 mW。
英文摘要：
Always-on edge cameras generate continuous video streams where redundant frames degrade cross-modal retrieval by crowding correct results out of top-k search. This paper presents a streaming retrieval architecture: an on-device epsilon-net filter retains only semantically novel frames, building a denoised embedding index; a cross-modal adapter and cloud re-ranker compensate for the compact encoder's weak alignment. A single-pass streaming filter outperforms offline alternatives (k-means, farthest-point, uniform, random) across eight vision-language models (8M-632M) on two egocentric datasets (AEA, EPIC-KITCHENS). Combined, the architecture reaches 45.6% Hit@5 on held-out data using an 8M on-device encoder at an estimated 2.7 mW.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
