Filters








5,999 Hits in 5.7 sec

Toward Semi-Supervised Graphical Object Detection in Document Images

Goutham Kallempudi, Khurram Azeem Hashmi, Alain Pagani, Marcus Liwicki, Didier Stricker, Muhammad Zeshan Afzal
2022 Future Internet  
This paper presents an end-to-end semi-supervised framework for graphical object detection in scanned document images to address this limitation.  ...  Moreover, our model trained on 10% of IIIT-AR-13K labeled data beats the previous fully supervised method +4.5 points.  ...  Proposed Framework Semi-supervised object detection algorithms are broadly divided into consistency and pseudo labeled.  ... 
doi:10.3390/fi14060176 fatcat:bmls5pqww5gorbgpc4s3kjtdwa

Exploiting Unlabeled Data with Vision and Language Models for Object Detection [article]

Shiyu Zhao, Zhixing Zhang, Samuel Schulter, Long Zhao, Vijay Kumar B.G, Anastasis Stathopoulos, Manmohan Chandraker, Dimitris Metaxas
2022 arXiv   pre-print
We demonstrate the value of the generated pseudo labels in two specific tasks, open-vocabulary detection, where a model needs to generalize to unseen object categories, and semi-supervised object detection  ...  Our empirical evaluation shows the effectiveness of the pseudo labels in both tasks, where we outperform competitive baselines and achieve a novel state-of-the-art for open-vocabulary object detection.  ...  Overview of pseudo labels (PLs) fusion for semi-supervised object detection (SSOD).  ... 
arXiv:2207.08954v1 fatcat:nfclx3fz3feezoxvici4ljijii

Curved Text Detection in Natural Scene Images with Semi- and Weakly-Supervised Learning [article]

Xugong Qin, Yu Zhou, Dongbao Yang, Weiping Wang
2019 arXiv   pre-print
A novel strategy which utilizes ground-truth bounding boxes to generate pseudo mask annotations is proposed in weakly-supervised learning.  ...  We propose a novel scheme to train an accurate text detector using only a small amount of pixel-level annotated data and a large amount of data annotated with rectangles or even unlabeled data.  ...  Illustration of the structure of the curved text detector [14] . • First, we propose a semi-and weakly-supervised curved text detection framework in which an accurate scene text detector is trained by  ... 
arXiv:1908.09990v1 fatcat:vk4qicccg5bvpkdtogaolsdib4

Ensemble-based Semi-Supervised Learning for Hate Speech Detection

Safa Alsafari
2021 Proceedings of the ... International Florida Artificial Intelligence Research Society Conference  
We assess these strategies by re-training all the classifiers with the seed dataset augmented with the trusted pseudo-labeled data.  ...  This paper introduces an ensemble-based semi-supervised learning approach to leverage the availability of abundant social media content.  ...  Selection of Confident Pseudo-abels In semi-supervised learning, pseudo-labels are utilized to optimize supervised models.  ... 
doi:10.32473/flairs.v34i1.128427 fatcat:3s7vttew45deneiernrzzypzi4

Improving problem detection in peer assessment through pseudo-labeling using semi-supervised learning

Chengyuan Liu, Jialin Cui, Ruixuan Shang, Yunkai Xiao, Qinjin Jia, Edward Gehringer, Antonija Mitrovic, Nigel Bosch
2022 Zenodo  
In this study, we propose to apply pseudo-labeling, a semi-supervised learning-based strategy, to improve the recognition of reviews that detect problems in the reviewed work.  ...  This is done by utilizing a small, reliably labeled dataset along with a large unlabeled dataset to train a text classifier.  ...  Pseudo-labeling is one of the most effective and efficient methods in semi-supervised learning [9] .  ... 
doi:10.5281/zenodo.6853126 fatcat:noxzz2iy3fg7vmsa7ruffd6c34

A novel self-learning semi-supervised deep learning network to detect fake news on social media

Xin Li, Peixin Lu, Lianting Hu, XiaoGuang Wang, Long Lu
2021 Multimedia tools and applications  
In order to address this issue, we designed a self-learning semi-supervised deep learning network by adding a confidence network layer, which made it possible to automatically return and add correct results  ...  Despite many existing fake news datasets, comprehensive and effective algorithms for detecting fake news have become one of the major obstacles.  ...  In this paper, we designed a self-learning semi-supervised deep learning network to detect fake news on social media.  ... 
doi:10.1007/s11042-021-11065-x pmid:34093070 pmcid:PMC8170457 fatcat:ccfrshrlmzdgtehxxikhtvkrhe

Pseudo-labelling Enhanced Media Bias Detection [article]

Qin Ruan, Brian Mac Namee, Ruihai Dong
2021 arXiv   pre-print
This paper proposes a simple but effective data augmentation method, which leverages the idea of pseudo-labelling to select samples from noisy distant supervision annotation datasets.  ...  Leveraging unlabelled data through weak or distant supervision is a compelling approach to developing more effective text classification models.  ...  Acknowledgement This publication has emanated from research conducted with the financial support of Science Foundation Ireland under Grant number 18/CRT/6183.  ... 
arXiv:2107.07705v1 fatcat:k75pheczdrgrdk2dzc54rzt3si

I3CL:Intra- and Inter-Instance Collaborative Learning for Arbitrary-shaped Scene Text Detection [article]

Bo Du, Jian Ye, Jing Zhang, Juhua Liu, Dacheng Tao
2022 arXiv   pre-print
Besides, to make full use of the unlabeled data, we design an effective semi-supervised learning method to leverage the pseudo labels via an ensemble strategy.  ...  text instances with diverse background context.  ...  Semi-supervised Learning Self-training using pseudo labels is a learning paradigm associated with constructing models in semi-supervised learning, which leverages the model's own confident predictions  ... 
arXiv:2108.01343v3 fatcat:l5fathqrvnai5m75hqe72hl23y

Penalizing Proposals using Classifiers for Semi-Supervised Object Detection [article]

Somnath Hazra, Pallab Dasgupta
2022 arXiv   pre-print
Semi-supervised object detection algorithms solve the problem with a small amount of gold-standard labels and a large unlabelled dataset used to generate silver-standard labels.  ...  In comparison with the baseline where no confidence metric is used, we achieved a 4% gain in mAP with 25% labeled data and 10% gain in mAP with 50% labeled data by using the proposed confidence metric.  ...  Semi-Supervised Object Detection Semi-supervised learning literature for images describes two kinds of methodologies.  ... 
arXiv:2205.13219v2 fatcat:pmd62ovza5bb5luavwfsbfdoci

Learning from Pixel-Level Label Noise: A New Perspective for Semi-Supervised Semantic Segmentation [article]

Rumeng Yi, Yaping Huang, Qingji Guan, Mengyang Pu, Runsheng Zhang
2021 arXiv   pre-print
This paper addresses semi-supervised semantic segmentation by exploiting a small set of images with pixel-level annotations (strong supervisions) and a large set of images with only image-level annotations  ...  In particular, for the generated pixel-level noisy labels from weak supervisions by Class Activation Map (CAM), we train a clean segmentation model with strong supervisions to detect the clean labels from  ...  The key step of semi-supervised methods is to infer accurate pixel-level labels for a large number of images with only image-level weak annotations.  ... 
arXiv:2103.14242v1 fatcat:kwhr3d5kgjbrlbvvnhrkd4jdsm

Neural Data-to-Text Generation Based on Small Datasets: Comparing the Added Value of Two Semi-Supervised Learning Approaches on Top of a Large Language Model [article]

Chris van der Lee, Thiago Castro Ferreira, Chris Emmery, Travis Wiltshire, Emiel Krahmer
2022 arXiv   pre-print
a pseudo-labeling semi-supervised learning approach.  ...  In terms of output quality, extending the training set of a data-to-text system with a language model using the pseudo-labeling approach did increase text quality scores, but the data augmentation approach  ...  Do the data augmentation and pseudo-label approaches differ in terms of output quality and output diversity when used as semi-supervised learning approaches in a neural data-to-text system with a language  ... 
arXiv:2207.06839v1 fatcat:uesa2x3xj5grjncf4pqi26mlnu

A Simple Semi-Supervised Learning Framework for Object Detection [article]

Kihyuk Sohn, Zizhao Zhang, Chun-Liang Li, Han Zhang, Chen-Yu Lee, Tomas Pfister
2020 arXiv   pre-print
We propose experimental protocols to evaluate the performance of semi-supervised object detection using MS-COCO and show the efficacy of STAC on both MS-COCO and VOC07.  ...  In this paper, we propose STAC, a simple yet effective SSL framework for visual object detection along with a data augmentation strategy.  ...  STAC trained with less accurate pseudo labels achieves only 24.25 mAP, while the one with more accurate pseudo labels achieves 30.30 mAP, confirming the importance of pseudo label quality.  ... 
arXiv:2005.04757v2 fatcat:qney5wxuebcc5dljxvdrf3lfzi

Reinforced Co-Training [article]

Jiawei Wu, Lei Li, William Yang Wang
2018 arXiv   pre-print
Experimental results on clickbait detection and generic text classification tasks demonstrate that our proposed method can obtain more accurate text classification results.  ...  Co-training is a popular semi-supervised learning framework to utilize a large amount of unlabeled data in addition to a small labeled set.  ...  (Dai and Le, 2015). • Semi-supervised CNN with Region Embedding (Region-SSL): The model learns embeddings of small text regions from unlabeled data for integration into a supervised CNN (Johnson and Zhang  ... 
arXiv:1804.06035v1 fatcat:w6kzq6wxo5fu5kvkgs423jlpnm

Reinforced Co-Training

Jiawei Wu, Lei Li, William Yang Wang
2018 Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers)  
Experimental results on clickbait detection and generic text classification tasks demonstrate that our proposed method can obtain more accurate text classification results.  ...  Co-training is a popular semi-supervised learning framework to utilize a large amount of unlabeled data in addition to a small labeled set.  ...  (Dai and Le, 2015). • Semi-supervised CNN with Region Embedding (Region-SSL): The model learns embeddings of small text regions from unlabeled data for integration into a supervised CNN (Johnson and Zhang  ... 
doi:10.18653/v1/n18-1113 dblp:conf/naacl/WuLW18 fatcat:72iawhfbd5evvju2c4kp6xtiny

Bias Bubbles: Using Semi-Supervised Learning to Measure How Many Biased News Articles Are Around Us

Qin Ruan, Brian Mac Namee, Ruihai Dong
2021 Irish Conference on Artificial Intelligence and Cognitive Science  
In this paper, we first explore the use of pseudo-labelling technology to mitigate this problem.  ...  Deep learning-based classifiers are one common way of identifying media bias, but they suffer from a lack of large-scale labelled datasets.  ...  This publication has emanated from research conducted with the financial support of Science Foundation Ireland under Grant number 18/CRT/6183.  ... 
dblp:conf/aics/RuanND21 fatcat:d5za3marpbg5dpcoycb3nzfray
« Previous Showing results 1 — 15 out of 5,999 results