arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2405.14730cs.CVcs.LG
收藏

用于高效重识别的嵌入压缩

Embedding Compression for Efficient Re-Identification

Modern Intelligence(现代智能公司);UC San Diego(加州大学圣地亚哥分校)

Luke McDermott

AI总结:

本文针对重识别系统存储嵌入规模受限的问题,评测了量化感知训练及三种降维方法,实现高达96倍压缩且性能损失极小,表明高维空间利用不足,为系统能力提升开辟了新方向。

摘要:

现实世界中的重识别(ReID)算法旨在将物体的新观测映射到先前记录的实例。这些系统通常受限于存储嵌入的数量和大小。为了解决这一扩展问题,我们尝试通过使用多种压缩技术来缩小这些向量的尺寸。在本文中,我们基准测试了量化感知训练以及三种不同的降维方法:迭代结构化剪枝、初始化时切片嵌入以及使用低秩嵌入。我们发现,ReID嵌入可以被压缩高达96倍,同时性能下降极小。这意味着现代重识别范式并未充分利用高维潜在空间,为进一步研究以增强这些系统的能力开辟了道路。

英文摘要:

Real world re-identfication (ReID) algorithms aim to map new observations of an object to previously recorded instances. These systems are often constrained by quantity and size of the stored embeddings. To combat this scaling problem, we attempt to shrink the size of these vectors by using a variety of compression techniques. In this paper, we benchmark quantization-aware-training along with three different dimension reduction methods: iterative structured pruning, slicing the embeddings at initialize, and using low rank embeddings. We find that ReID embeddings can be compressed by up to 96x with minimal drop in performance. This implies that modern re-identification paradigms do not fully leverage the high dimensional latent space, opening up further research to increase the capabilities of these systems.

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。

↑