Scholarlink

Abstract <p>We propose a two-stage, reliability-aware pipeline for video-based human attribute recognition. A&#xa0;pretrained vision–language model extracts attribute-relevant frame features, which a lightweight multilayer perceptron decodes into per-frame probabilities. To aggregate noisy frame scores into stable track-level outputs, we introduce the anchor-clustering aggregation method: probabilities are discretized into ordered anchors, and the final decision is based on anchor counts and the longest high-anchor streak, favoring temporal consistency over isolated confidence spikes. The proposed method yields relative improvements of 9.55 and 13.54% in True Positive Rate on the MARS-Attribute and DukeMTMC-VideoReID-Attribute datasets, respectively, while maintaining a high F1 score and minimal inference overhead of milliseconds. The proposed framework is also flexible and easy to adapt beyond human attribute recognition, making it applicable to a wide range of temporal aggregation tasks.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Anchor-Clustering: A Lightweight Efficient Method to Aggregate Human Attribute Prediction in Video Surveillance

  • 期刊论文

  • 发表日期：

    2026年3月1日

查看全文
  • Manh Tu Vu,
  • Hoang Nam Vu,
  • Hoang Long Nguyen

摘要

Abstract

We propose a two-stage, reliability-aware pipeline for video-based human attribute recognition. A pretrained vision–language model extracts attribute-relevant frame features, which a lightweight multilayer perceptron decodes into per-frame probabilities. To aggregate noisy frame scores into stable track-level outputs, we introduce the anchor-clustering aggregation method: probabilities are discretized into ordered anchors, and the final decision is based on anchor counts and the longest high-anchor streak, favoring temporal consistency over isolated confidence spikes. The proposed method yields relative improvements of 9.55 and 13.54% in True Positive Rate on the MARS-Attribute and DukeMTMC-VideoReID-Attribute datasets, respectively, while maintaining a high F1 score and minimal inference overhead of milliseconds. The proposed framework is also flexible and easy to adapt beyond human attribute recognition, making it applicable to a wide range of temporal aggregation tasks.