Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

BiSPL: Bidirectional Self-Paced Learning for recognition from web data

Wu, Xiaoping, Chang, Jianlong, Lai, Yu-Kun ORCID: https://orcid.org/0000-0002-2094-5680, Yang, Jufeng and Tian, Qi 2021. BiSPL: Bidirectional Self-Paced Learning for recognition from web data. IEEE Transactions on Image Processing 30 , 6512 - 6527. 10.1109/TIP.2021.3094744

[thumbnail of BiSPL_TIP.pdf] PDF - Accepted Post-Print Version
Download (5MB)

Abstract

Deep learning (DL) is inherently subject to the requirement of a large amount of well-labeled data, which is expensive and time-consuming to obtain manually. In order to broaden the reach of DL, leveraging free web data becomes an attractive strategy to alleviate the issue of data scarcity. However, directly utilizing collected web data to train a deep model is ineffective because of the mixed noisy data. To address such problems, we develop a novel bidirectional self-paced learning (BiSPL) framework which reduces the effect of noise by learning from web data in a meaningful order. Technically, the BiSPL framework consists of two essential steps. Relying on distances defined between web samples and labeled source samples, first, the web samples with short distances are sampled and combined to form a new training set. Second, based on the new training set, both easy and hard samples are initially employed to train deep models for higher stability, and hard samples are gradually dropped to reduce the noise as the training progresses. By iteratively alternating such steps, deep models converge to a better solution. We mainly focus on the fine-grained visual classification (FGVC) tasks because their corresponding datasets are generally small and therefore face a more significant data scarcity problem. Experiments conducted on six public FGVC tasks demonstrate that our proposed method outperforms the state-of-the-art approaches. Especially, BiSPL suffices to achieve the highest stable performance when the scale of the well-labeled training set decreases dramatically.

Item Type: Article
Date Type: Publication
Status: Published
Schools: Computer Science & Informatics
Publisher: Institute of Electrical and Electronics Engineers
ISSN: 1057-7149
Date of First Compliant Deposit: 30 July 2021
Date of Acceptance: 26 June 2021
Last Modified: 23 Nov 2022 20:59
URI: https://orca.cardiff.ac.uk/id/eprint/143075

Citation Data

Cited 1 time in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item Edit Item

Downloads

Downloads per month over past year

View more statistics