Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Ping, Wang, Hongbo, Lu, Lei
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2410.09797
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912122207731712
author Li, Ping
Wang, Hongbo
Lu, Lei
author_facet Li, Ping
Wang, Hongbo
Lu, Lei
contents Metric-based few-shot fine-grained classification has shown promise due to its simplicity and efficiency. However, existing methods often overlook task-level special cases and struggle with accurate category description and irrelevant sample information. To tackle these, we propose TAFD-Net: a task adaptive feature distribution network. It features a task-adaptive component for embedding to capture task-level nuances, an asymmetric metric for calculating feature distribution similarities between query samples and support categories, and a contrastive measure strategy to boost performance. Extensive experiments have been conducted on three datasets and the experimental results show that our proposed algorithm outperforms recent incremental learning algorithms.
format Preprint
id arxiv_https___arxiv_org_abs_2410_09797
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Task Adaptive Feature Distribution Based Network for Few-shot Fine-grained Target Classification
Li, Ping
Wang, Hongbo
Lu, Lei
Computer Vision and Pattern Recognition
Metric-based few-shot fine-grained classification has shown promise due to its simplicity and efficiency. However, existing methods often overlook task-level special cases and struggle with accurate category description and irrelevant sample information. To tackle these, we propose TAFD-Net: a task adaptive feature distribution network. It features a task-adaptive component for embedding to capture task-level nuances, an asymmetric metric for calculating feature distribution similarities between query samples and support categories, and a contrastive measure strategy to boost performance. Extensive experiments have been conducted on three datasets and the experimental results show that our proposed algorithm outperforms recent incremental learning algorithms.
title Task Adaptive Feature Distribution Based Network for Few-shot Fine-grained Target Classification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.09797