13,243 Hits in 5.4 sec

Improving Object Detection with Selective Self-supervised Self-training [article]

Yandong Li, Di Huang, Danfeng Qin, Liqiang Wang, Boqing Gong
2020 arXiv   pre-print
On the other hand, we propose a novel learning method motivated by two parallel lines of work that explore unlabeled data for image classification: self-training and self-supervised learning.  ...  The Web images are diverse, supplying a wide variety of object poses, appearances, their interactions with the context, etc.  ...  Hence, we build our approach upon self-training in this paper. Semi-supervised Object Detection.  ... 
arXiv:2007.09162v2 fatcat:jubt7f4alje47ghb3c7jntpyfa

One Million Scenes for Autonomous Driving: ONCE Dataset [article]

Jiageng Mao, Minzhe Niu, Chenhan Jiang, Hanxue Liang, Jingheng Chen, Xiaodan Liang, Yamin Li, Chaoqiang Ye, Wei Zhang, Zhenguo Li, Jie Yu, Hang Xu (+1 others)
2021 arXiv   pre-print
To facilitate future research on exploiting unlabeled data for 3D detection, we additionally provide a benchmark in which we reproduce and evaluate a variety of self-supervised and semi-supervised methods  ...  In this paper, we introduce the ONCE (One millioN sCenEs) dataset for 3D object detection in the autonomous driving scenario.  ...  Mean Teacher uses the teacher and student model for semi-supervised learning.  ... 
arXiv:2106.11037v3 fatcat:fwgrb57yarhujmetzpewtdzzei

A Multi-task Mean Teacher for Semi-supervised Facial Affective Behavior Analysis [article]

Lingfeng Wang, Shisen Wang, Jin Qi, Kenji Suzuki
2021 arXiv   pre-print
To tackle this problem, this paper presents a semi-supervised model with a mean teacher framework to leverage additional unlabeled data.  ...  When training with unlabeled data, the teacher network is employed to predict pseudo labels for student network training, which allows it to learn from unlabeled data.  ...  Then we employ the mean teacher with semi-supervised learning to learn from additional unlabeled data for further improving the recognition performance.  ... 
arXiv:2107.04225v3 fatcat:qnqnloebtbgz3mihjv6ppqctpa

From Handheld to Unconstrained Object Detection: a Weakly-supervised On-line Learning Approach [article]

Elisa Maiettini and Andrea Maracani and Raffaello Camoriano and Giulia Pasquale and Vadim Tikhanoff and Lorenzo Rosasco and Lorenzo Natale
2022 arXiv   pre-print
We show that the robot can improve adaptation to novel domains, either by interacting with a human teacher (Active Learning) or with an autonomous supervision (Semi-supervised Learning).  ...  In this work, we target the scenario of a robot trained in a teacher-learner setting to detect handheld objects.  ...  A first detection model is trained during a brief interaction with a human, in a teacher-learner scenario, like e.g. in [4] where objects are handheld (the TARGET-LABELED).  ... 
arXiv:2012.14345v2 fatcat:7uzjskvbkzefpiw22ywhmjfjbi

Scale-Equivalent Distillation for Semi-Supervised Object Detection [article]

Qiushan Guo, Yao Mu, Jianyu Chen, Tianqi Wang, Yizhou Yu, Ping Luo
2022 arXiv   pre-print
Recent Semi-Supervised Object Detection (SS-OD) methods are mainly based on self-training, i.e., generating hard pseudo-labels by a teacher model on unlabeled data as supervisory signals.  ...  Although they achieved certain success, the limited labeled data in semi-supervised learning scales up the challenges of object detection.  ...  However, Semi-Supervised Object Detection is more challenging than Semi-Supervised Image Classification on the balanced dataset.  ... 
arXiv:2203.12244v2 fatcat:otalw2k6bbcgnhtctco65txv6y

A semi-supervised learning detection method for vision-based monitoring of construction sites by integrating teacher-student networks and data augmentation

Bo Xiao, Yuxuan Zhang, Yuan Chen, Xianfei Yin
2021 Advanced Engineering Informatics  
To address this problem, this research proposes a semi-supervised learning detection method for construction site monitoring based on teacher-student networks and data augmentation.  ...  Initially, the proposed method trains the teacher object detector with labeled data following weak data augmentation.  ...  Acknowledgement The authors would like to thank Professor Shih-Chung Kang for publishing the ACID dataset to the community.  ... 
doi:10.1016/j.aei.2021.101372 fatcat:wxkw7xz4lfghrkwiq462ccj3mu

Medical Image Segmentation with Limited Supervision: A Review of Deep Network Models [article]

Jialin Peng, Ye Wang
2021 arXiv   pre-print
However, due to its intrinsic difficulty, segmentation with limited supervision is challenging and specific model design and/or learning strategies are needed.  ...  Despite the remarkable performance of deep learning methods on various tasks, most cutting-edge models rely heavily on large-scale annotated training examples, which are often unavailable for clinical  ...  [180] adapted the mean teacher model [151] , an improved teacher-student self-training strategy that also considered consistency regularization, to semi-supervised brain lesion segmentation.  ... 
arXiv:2103.00429v1 fatcat:p44a5e34sre4nasea5kjvva55e

Medical Image Segmentation with Limited Supervision: A Review of Deep Network Models

Jialin Peng, Ye Wang
2021 IEEE Access  
INDEX TERMS Medical image segmentation, semi-supervised segmentation, partially-supervised segmentation, noisy label, sparse annotation. 36828  ...  Despite the remarkable performance of deep learning methods on various tasks, most cutting-edge models rely heavily on large-scale annotated training examples, which are often unavailable for clinical  ...  [180] adapted the mean teacher model [151] , an improved teacher-student self-training strategy that also considered consistency regularization, to semi-supervised brain lesion segmentation.  ... 
doi:10.1109/access.2021.3062380 fatcat:r5vsec2yfzcy5nk7wusiftyayu

Unbiased Teacher for Semi-Supervised Object Detection [article]

Yen-Cheng Liu, Chih-Yao Ma, Zijian He, Chia-Wen Kuo, Kan Chen, Peizhao Zhang, Bichen Wu, Zsolt Kira, Peter Vajda
2021 arXiv   pre-print
Semi-supervised learning, i.e., training networks with both labeled and unlabeled data, has made significant progress recently.  ...  In this work, we revisit the Semi-Supervised Object Detection (SS-OD) and identify the pseudo-labeling bias issue in SS-OD.  ...  Semi-Supervised Object Detection.  ... 
arXiv:2102.09480v1 fatcat:cmmalphe2ndgdmxpmqdmy2uvme

Tackling the Problem of Limited Data and Annotations in Semantic Segmentation [article]

Ahmadreza Jeddi
2020 arXiv   pre-print
To this end, RotNet, DeeperCluster, and Semi&Weakly Supervised Learning (SWSL) pre-trained models are transferred and finetuned in a DeepLab-v2 baseline, and dense CRF is applied both as a post-processing  ...  In this work, the case of semantic segmentation on a small image dataset (simulated by 1000 randomly selected images from PASCAL VOC 2012), where only weak supervision signals (scribbles from user interaction  ...  Semi-Weakly Supervised ImageNet Model In [46] , Yalniz et al. train an ImageNet classifier using a teacher-student method as follows: first, a powerful ImageNet classifier is trained as the teacher model  ... 
arXiv:2007.07357v1 fatcat:ftd5uszayvhyppqlaqblxlc4rm

Unsupervised Anomaly Detection with Distillated Teacher-Student Network Ensemble

Qinfeng Xiao, Jing Wang, Youfang Lin, Wenbo Gongsa, Ganghui Hu, Menggang Li, Fang Wang
2021 Entropy  
To effectively detect anomalies for multivariate data, this paper introduces a teacher-student distillation based framework Distillated Teacher-Student Network Ensemble (DTSNE).  ...  The paradigm of the teacher-student distillation is able to deal with high-dimensional complex features.  ...  Self-supervised Pre-training. Compared with using a random projection like [25] , we pre-train the teacher network with a self-supervised objective.  ... 
doi:10.3390/e23020201 pmid:33561954 pmcid:PMC7915583 fatcat:btj5sdjbwzgenpvbnrlrkcctae

Semi-supervised Object Detection via Virtual Category Learning [article]

Changrui Chen, Kurt Debattista, Jungong Han
2022 arXiv   pre-print
Due to the costliness of labelled data in real-world applications, semi-supervised object detectors, underpinned by pseudo labelling, are appealing.  ...  However, handling confusing samples is nontrivial: discarding valuable confusing samples would compromise the model generalisation while using them for training would exacerbate the confirmation bias issue  ...  Semi-supervised Object Detection (SSOD) originates from semi-supervised classification, where only a small amount of bounding box labelled data and numerous unlabelled data are available for training a  ... 
arXiv:2207.03433v1 fatcat:d72wuoiftrfv7ltg7vho5zejpi

Exploring Object Relation in Mean Teacher for Cross-Domain Detection [article]

Qi Cai, Yingwei Pan, Chong-Wah Ngo, Xinmei Tian, Lingyu Duan, Ting Yao
2019 arXiv   pre-print
Specifically, we present Mean Teacher with Object Relations (MTOR) that novelly remolds Mean Teacher under the backbone of Faster R-CNN by integrating the object relations into the measure of consistency  ...  The domain gap is thus naturally bridged with consistency regularization in a teacher-student scheme. In this work, we advance this Mean Teacher paradigm to be applicable for cross-domain detection.  ...  Mean Teacher in Semi-Supervised Learning We briefly review semi-supervised learning with Mean Teacher [48] .  ... 
arXiv:1904.11245v2 fatcat:ymy6aajapvfb7ohytndmtiue7e

Label Noise-Resistant Mean Teaching for Weakly Supervised Fake News Detection [article]

Jingyi Xie, Jiawei Liu, Zheng-Jun Zha
2022 arXiv   pre-print
In this work, we propose a novel label noise-resistant mean teaching approach (LNMT) for weakly supervised fake news detection.  ...  Moreover, in order to suppress the noises in weak labels, LNMT establishes a mean teacher framework equipped with label propagation and label reliability estimation.  ...  Moreover, LNMT establishes a mean teacher framework with different learning ability and information interaction.  ... 
arXiv:2206.12260v1 fatcat:546wczwh25b6hbh2jdbraxnidi

DetMatch: Two Teachers are Better Than One for Joint 2D and 3D Semi-Supervised Object Detection [article]

Jinhyung Park, Chenfeng Xu, Yiyang Zhou, Masayoshi Tomizuka, Wei Zhan
2022 arXiv   pre-print
Observing that the distinct characteristics of each sensor cause them to be biased towards detecting different objects, we propose DetMatch, a flexible framework for joint semi-supervised learning on 2D  ...  While numerous 3D detection works leverage the complementary relationship between RGB images and point clouds, developments in the broader framework of semi-supervised object recognition remain uninfluenced  ...  After the student is pre-trained to convergence, the teacher is initialized with the student weights before the SSL training begins. Semi-Supervised Training.  ... 
arXiv:2203.09510v1 fatcat:ryx6s3bs4ndftg3y3iaat6grcm
« Previous Showing results 1 — 15 out of 13,243 results