Storing Less, Finding More: How Novelty Filtering Improves Cross-Modal Retrieval on Edge Cameras
This addresses the challenge of efficient and accurate cross-modal retrieval for edge camera applications, representing an incremental improvement over existing methods.
The paper tackles the problem of redundant frames degrading cross-modal retrieval in continuous video streams from edge cameras by introducing a streaming retrieval architecture with an on-device novelty filter, achieving 45.6% Hit@5 on held-out data using an 8M encoder at 2.7 mW.
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.