Proto-OOD: Enhancing OOD Object Detection with Prototype Feature Similarity

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Junkun, Mei, Jilin, Chen, Liang, Zhao, Fangzhou, Xing, Yan, Hu, Yu
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913667371499520
author Chen, Junkun
Mei, Jilin
Chen, Liang
Zhao, Fangzhou
Xing, Yan
Hu, Yu
author_facet Chen, Junkun
Mei, Jilin
Chen, Liang
Zhao, Fangzhou
Xing, Yan
Hu, Yu
contents Neural networks that are trained on limited category samples often mispredict out-of-distribution (OOD) objects. We observe that features of the same category are more tightly clustered in feature space, while those of different categories are more dispersed. Based on this, we propose using prototype similarity for OOD detection. Drawing on widely used prototype features in few-shot learning, we introduce a novel OOD detection network structure (Proto-OOD). Proto-OOD enhances the representativeness of category prototypes using contrastive loss and detects OOD data by evaluating the similarity between input features and category prototypes. During training, Proto-OOD generates OOD samples for training the similarity module with a negative embedding generator. When Pascal VOC are used as the in-distribution dataset and MS-COCO as the OOD dataset, Proto-OOD significantly reduces the FPR (false positive rate). Moreover, considering the limitations of existing evaluation metrics, we propose a more reasonable evaluation protocol. The code will be released.
format Preprint
id arxiv_https___arxiv_org_abs_2409_05466
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Proto-OOD: Enhancing OOD Object Detection with Prototype Feature Similarity
Chen, Junkun
Mei, Jilin
Chen, Liang
Zhao, Fangzhou
Xing, Yan
Hu, Yu
Computer Vision and Pattern Recognition
Artificial Intelligence
Neural networks that are trained on limited category samples often mispredict out-of-distribution (OOD) objects. We observe that features of the same category are more tightly clustered in feature space, while those of different categories are more dispersed. Based on this, we propose using prototype similarity for OOD detection. Drawing on widely used prototype features in few-shot learning, we introduce a novel OOD detection network structure (Proto-OOD). Proto-OOD enhances the representativeness of category prototypes using contrastive loss and detects OOD data by evaluating the similarity between input features and category prototypes. During training, Proto-OOD generates OOD samples for training the similarity module with a negative embedding generator. When Pascal VOC are used as the in-distribution dataset and MS-COCO as the OOD dataset, Proto-OOD significantly reduces the FPR (false positive rate). Moreover, considering the limitations of existing evaluation metrics, we propose a more reasonable evaluation protocol. The code will be released.
title Proto-OOD: Enhancing OOD Object Detection with Prototype Feature Similarity
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2409.05466