End-to-End: A Simple Template for the Long-Tailed-Recognition of Transmission Line Clamps via a Vision-Language Model
Raw image classification datasets generally maintain a long-tailed distribution in the real world. Standard classification algorithms face a substantial issue because many labels only relate to a few categories. The model learning processes will tend toward the dominant labels under the influence of...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-03-01
|
Series: | Applied Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3417/13/5/3287 |
_version_ | 1797615630373879808 |
---|---|
author | Fei Yan Hui Zhang Yaogen Li Yongjia Yang Yinping Liu |
author_facet | Fei Yan Hui Zhang Yaogen Li Yongjia Yang Yinping Liu |
author_sort | Fei Yan |
collection | DOAJ |
description | Raw image classification datasets generally maintain a long-tailed distribution in the real world. Standard classification algorithms face a substantial issue because many labels only relate to a few categories. The model learning processes will tend toward the dominant labels under the influence of their loss functions. Existing systems typically use two stages to improve performance: pretraining on initial imbalanced datasets and fine-tuning on balanced datasets via re-sampling or logit adjustment. These have achieved promising results. However, their limited self-supervised information makes it challenging to transfer such systems to other vision tasks, such as detection and segmentation. Using large-scale contrastive visual-language pretraining, the Open AI team discovered a novel visual recognition method. We provide a simple one-stage model called the text-to-image network (TIN) for long-tailed recognition (LTR) based on the similarities between textual and visual features. The TIN has the following advantages over existing techniques: (1) Our model incorporates textual and visual semantic information. (2) This end-to-end strategy achieves good results with fewer image samples and no secondary training. (3) By using seesaw loss, we further reduce the loss gap between the head category and the tail category. These adjustments encourage large relative magnitudes between the logarithms of rare and dominant labels. TIN conducted extensive comparative experiments with a large number of advanced models on ImageNet-LT, the largest long-tailed public dataset, and achieved the state-of-the-art for a single-stage model with 72.8% at Top-1 accuracy. |
first_indexed | 2024-03-11T07:29:24Z |
format | Article |
id | doaj.art-0feb2911ac2440688b4135e8380d1cde |
institution | Directory Open Access Journal |
issn | 2076-3417 |
language | English |
last_indexed | 2024-03-11T07:29:24Z |
publishDate | 2023-03-01 |
publisher | MDPI AG |
record_format | Article |
series | Applied Sciences |
spelling | doaj.art-0feb2911ac2440688b4135e8380d1cde2023-11-17T07:21:48ZengMDPI AGApplied Sciences2076-34172023-03-01135328710.3390/app13053287End-to-End: A Simple Template for the Long-Tailed-Recognition of Transmission Line Clamps via a Vision-Language ModelFei Yan0Hui Zhang1Yaogen Li2Yongjia Yang3Yinping Liu4College of Automation, Nanjing University of Information Science & Technology, Nanjing 210044, ChinaCollege of Automation, Nanjing University of Information Science & Technology, Nanjing 210044, ChinaCollege of Automation, Nanjing University of Information Science & Technology, Nanjing 210044, ChinaCollege of Automation, Nanjing University of Information Science & Technology, Nanjing 210044, ChinaCollege of Automation, Nanjing University of Information Science & Technology, Nanjing 210044, ChinaRaw image classification datasets generally maintain a long-tailed distribution in the real world. Standard classification algorithms face a substantial issue because many labels only relate to a few categories. The model learning processes will tend toward the dominant labels under the influence of their loss functions. Existing systems typically use two stages to improve performance: pretraining on initial imbalanced datasets and fine-tuning on balanced datasets via re-sampling or logit adjustment. These have achieved promising results. However, their limited self-supervised information makes it challenging to transfer such systems to other vision tasks, such as detection and segmentation. Using large-scale contrastive visual-language pretraining, the Open AI team discovered a novel visual recognition method. We provide a simple one-stage model called the text-to-image network (TIN) for long-tailed recognition (LTR) based on the similarities between textual and visual features. The TIN has the following advantages over existing techniques: (1) Our model incorporates textual and visual semantic information. (2) This end-to-end strategy achieves good results with fewer image samples and no secondary training. (3) By using seesaw loss, we further reduce the loss gap between the head category and the tail category. These adjustments encourage large relative magnitudes between the logarithms of rare and dominant labels. TIN conducted extensive comparative experiments with a large number of advanced models on ImageNet-LT, the largest long-tailed public dataset, and achieved the state-of-the-art for a single-stage model with 72.8% at Top-1 accuracy.https://www.mdpi.com/2076-3417/13/5/3287unmanned aerial vehiclestate gridtransmission line clampsimage classificationmultimodule fusionneural network |
spellingShingle | Fei Yan Hui Zhang Yaogen Li Yongjia Yang Yinping Liu End-to-End: A Simple Template for the Long-Tailed-Recognition of Transmission Line Clamps via a Vision-Language Model Applied Sciences unmanned aerial vehicle state grid transmission line clamps image classification multimodule fusion neural network |
title | End-to-End: A Simple Template for the Long-Tailed-Recognition of Transmission Line Clamps via a Vision-Language Model |
title_full | End-to-End: A Simple Template for the Long-Tailed-Recognition of Transmission Line Clamps via a Vision-Language Model |
title_fullStr | End-to-End: A Simple Template for the Long-Tailed-Recognition of Transmission Line Clamps via a Vision-Language Model |
title_full_unstemmed | End-to-End: A Simple Template for the Long-Tailed-Recognition of Transmission Line Clamps via a Vision-Language Model |
title_short | End-to-End: A Simple Template for the Long-Tailed-Recognition of Transmission Line Clamps via a Vision-Language Model |
title_sort | end to end a simple template for the long tailed recognition of transmission line clamps via a vision language model |
topic | unmanned aerial vehicle state grid transmission line clamps image classification multimodule fusion neural network |
url | https://www.mdpi.com/2076-3417/13/5/3287 |
work_keys_str_mv | AT feiyan endtoendasimpletemplateforthelongtailedrecognitionoftransmissionlineclampsviaavisionlanguagemodel AT huizhang endtoendasimpletemplateforthelongtailedrecognitionoftransmissionlineclampsviaavisionlanguagemodel AT yaogenli endtoendasimpletemplateforthelongtailedrecognitionoftransmissionlineclampsviaavisionlanguagemodel AT yongjiayang endtoendasimpletemplateforthelongtailedrecognitionoftransmissionlineclampsviaavisionlanguagemodel AT yinpingliu endtoendasimpletemplateforthelongtailedrecognitionoftransmissionlineclampsviaavisionlanguagemodel |