Infinite-dimensional feature aggregation via a factorized bilinear model

作者：

Highlights：

• Infinite-dimensional features are directly aggregated without approximation error.

• Our descriptors contain infinite order statistics of input features.

• The sigmoid kernel is introduced to construct infinite-dimensional features.

• Our method outperforms the state-of-the-art finite-dimensional and infinite-dimensional feature aggregation methods.

摘要

•Infinite-dimensional features are directly aggregated without approximation error.•Our descriptors contain infinite order statistics of input features.•The sigmoid kernel is introduced to construct infinite-dimensional features.•Our method outperforms the state-of-the-art finite-dimensional and infinite-dimensional feature aggregation methods.

论文关键词：Feature aggregation,Infinite-dimensional features,Non-approximate method,Second-order statistics

论文评审过程：Received 17 August 2020, Revised 8 July 2021, Accepted 19 October 2021, Available online 20 November 2021, Version of Record 30 December 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108397