Saved in:
Bibliographic Details
Main Authors: Park, Junwoo, Lee, Jangho, Lim, Sunho
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.11714
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913092972052480
author Park, Junwoo
Lee, Jangho
Lim, Sunho
author_facet Park, Junwoo
Lee, Jangho
Lim, Sunho
contents Pretrained detectors perform well on benchmarks but often suffer performance degradation in real-world deployments due to distribution gaps between training data and target environments. COCO-like benchmarks emphasize category diversity rather than instance density, causing detectors trained under per-class sparsity to struggle in dense, single- or few-class scenes such as surveillance and traffic monitoring. In fixed-camera environments, the quasi-static background provides a stable, label-free prior that can be exploited at inference to suppress spurious detections. To address the issue, we propose Background Embedding Memory (BEM), a lightweight, training-free, weight-frozen module that can be attached to pretrained detectors during inference. BEM estimates clean background embeddings, maintains a prototype memory, and re-scores detection logits with an inverse-similarity, rank-weighted penalty, effectively reducing false positives while maintaining recall. Empirically, background-frame cosine similarity correlates negatively with object count and positively with Precision-Confidence AUC (P-AUC), motivating its use as a training-free control signal. Across YOLO and RT-DETR families on LLVIP and simulated surveillance streams, BEM consistently reduces false positives while preserving real-time performance. Our code is available at https://github.com/Leo-Park1214/Background-Embedding-Memory.git
format Preprint
id arxiv_https___arxiv_org_abs_2604_11714
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle BEM: Training-Free Background Embedding Memory for False-Positive Suppression in Real-Time Fixed-Background Camera
Park, Junwoo
Lee, Jangho
Lim, Sunho
Computer Vision and Pattern Recognition
Pretrained detectors perform well on benchmarks but often suffer performance degradation in real-world deployments due to distribution gaps between training data and target environments. COCO-like benchmarks emphasize category diversity rather than instance density, causing detectors trained under per-class sparsity to struggle in dense, single- or few-class scenes such as surveillance and traffic monitoring. In fixed-camera environments, the quasi-static background provides a stable, label-free prior that can be exploited at inference to suppress spurious detections. To address the issue, we propose Background Embedding Memory (BEM), a lightweight, training-free, weight-frozen module that can be attached to pretrained detectors during inference. BEM estimates clean background embeddings, maintains a prototype memory, and re-scores detection logits with an inverse-similarity, rank-weighted penalty, effectively reducing false positives while maintaining recall. Empirically, background-frame cosine similarity correlates negatively with object count and positively with Precision-Confidence AUC (P-AUC), motivating its use as a training-free control signal. Across YOLO and RT-DETR families on LLVIP and simulated surveillance streams, BEM consistently reduces false positives while preserving real-time performance. Our code is available at https://github.com/Leo-Park1214/Background-Embedding-Memory.git
title BEM: Training-Free Background Embedding Memory for False-Positive Suppression in Real-Time Fixed-Background Camera
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.11714