BackMix: Regularizing Open Set Recognition by Removing Underlying Fore-Background Priors

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Yu, Mu, Junxian, Huang, Hongzhi, Wang, Qilong, Zhu, Pengfei, Hu, Qinghua
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917966409367552
author Wang, Yu
Mu, Junxian
Huang, Hongzhi
Wang, Qilong
Zhu, Pengfei
Hu, Qinghua
author_facet Wang, Yu
Mu, Junxian
Huang, Hongzhi
Wang, Qilong
Zhu, Pengfei
Hu, Qinghua
contents Open set recognition (OSR) requires models to classify known samples while detecting unknown samples for real-world applications. Existing studies show impressive progress using unknown samples from auxiliary datasets to regularize OSR models, but they have proved to be sensitive to selecting such known outliers. In this paper, we discuss the aforementioned problem from a new perspective: Can we regularize OSR models without elaborately selecting auxiliary known outliers? We first empirically and theoretically explore the role of foregrounds and backgrounds in open set recognition and disclose that: 1) backgrounds that correlate with foregrounds would mislead the model and cause failures when encounters 'partially' known images; 2) Backgrounds unrelated to foregrounds can serve as auxiliary known outliers and provide regularization via global average pooling. Based on the above insights, we propose a new method, Background Mix (BackMix), that mixes the foreground of an image with different backgrounds to remove the underlying fore-background priors. Specifically, BackMix first estimates the foreground with class activation maps (CAMs), then randomly replaces image patches with backgrounds from other images to obtain mixed images for training. With backgrounds de-correlated from foregrounds, the open set recognition performance is significantly improved. The proposed method is quite simple to implement, requires no extra operation for inferences, and can be seamlessly integrated into almost all of the existing frameworks. The code is released on https://github.com/Vanixxz/BackMix.
format Preprint
id arxiv_https___arxiv_org_abs_2503_17717
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BackMix: Regularizing Open Set Recognition by Removing Underlying Fore-Background Priors
Wang, Yu
Mu, Junxian
Huang, Hongzhi
Wang, Qilong
Zhu, Pengfei
Hu, Qinghua
Computer Vision and Pattern Recognition
Open set recognition (OSR) requires models to classify known samples while detecting unknown samples for real-world applications. Existing studies show impressive progress using unknown samples from auxiliary datasets to regularize OSR models, but they have proved to be sensitive to selecting such known outliers. In this paper, we discuss the aforementioned problem from a new perspective: Can we regularize OSR models without elaborately selecting auxiliary known outliers? We first empirically and theoretically explore the role of foregrounds and backgrounds in open set recognition and disclose that: 1) backgrounds that correlate with foregrounds would mislead the model and cause failures when encounters 'partially' known images; 2) Backgrounds unrelated to foregrounds can serve as auxiliary known outliers and provide regularization via global average pooling. Based on the above insights, we propose a new method, Background Mix (BackMix), that mixes the foreground of an image with different backgrounds to remove the underlying fore-background priors. Specifically, BackMix first estimates the foreground with class activation maps (CAMs), then randomly replaces image patches with backgrounds from other images to obtain mixed images for training. With backgrounds de-correlated from foregrounds, the open set recognition performance is significantly improved. The proposed method is quite simple to implement, requires no extra operation for inferences, and can be seamlessly integrated into almost all of the existing frameworks. The code is released on https://github.com/Vanixxz/BackMix.
title BackMix: Regularizing Open Set Recognition by Removing Underlying Fore-Background Priors
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.17717