Filters








19,051 Hits in 4.6 sec

Restoring Negative Information in Few-Shot Object Detection [article]

Yukuan Yang, Fangyun Wei, Miaojing Shi, Guoqi Li
2020 arXiv   pre-print
In this paper, we restore the negative information in few-shot object detection by introducing a new negative- and positive-representative based metric learning framework and a new inference scheme with  ...  Negatives, especially hard negatives, however, are essential to the embedding space learning in few-shot object detection.  ...  In this paper, we propose to restore the negative information in the few-shot object detection: we show that hard negatives are essential for the metric learning in few-shot object detection.  ... 
arXiv:2010.11714v2 fatcat:fpyly7rbyzdnrcjxlp4w337wam

Insulator Anomaly Detection Method Based on Few-Shot Learning

Zhaoyang Wang, Qiang Gao, Dong Li, Junjie Liu, Hongwei Wang, Xiao Yu, Yipin Wang
2021 IEEE Access  
few-shot object detection model.  ...  In this stage, we build a few-shot object detection model based on the multi-scale feature reweighting network and detect anomaly insulator caps on the image cropped in the first stage. 1) FEW-SHOT LEARNING  ... 
doi:10.1109/access.2021.3071305 fatcat:5ahx2a4gunbe3i6jeacjssjc3q

Few-Shot Object Detection with Proposal Balance Refinement [article]

Sueyeon Kim, Woo-Jeoung Nam, Seong-Whan Lee
2022 arXiv   pre-print
Few-shot object detection has gained significant attention in recent years as it has the potential to greatly reduce the reliance on large amounts of manually annotated bounding boxes.  ...  To alleviate this issue, we present a few-shot object detection model with proposal balance refinement, a simple yet effective approach in learning object proposals using an auxiliary sequential bounding  ...  representatives by restoring negative information.  ... 
arXiv:2204.10527v1 fatcat:b7e5gcuut5emfldt2dieaskvxi

Hallucination Improves Few-Shot Object Detection [article]

Weilin Zhang, Yu-Xiong Wang
2021 arXiv   pre-print
One critical factor in improving few-shot detection is to address the lack of variation in training data.  ...  Learning to detect novel objects from few annotated examples is of great practical importance.  ...  Few-Shot Object Detection: Advanced few-shot detectors are usually built in a serial fashion [7, 38, 39, 43, 44, 45] .  ... 
arXiv:2105.01294v1 fatcat:pl6664zeqrasljjoj5hq5cq6ny

Few-shot Object Detection via Sample Processing

Honghui Xu, Xinqing Wang, Faming Shao, Baoguo Duan, Peng Zhang
2021 IEEE Access  
ACKNOWLEDGMENT The authors would like to thank the editor-in-chief, the associate editor, and the reviewers for their insightful comments and suggestions.  ...  few-shot object detection.  ...  In this paper, a novel few-shot object detection model is proposed to address the above dilemma.  ... 
doi:10.1109/access.2021.3059446 fatcat:2hjmabag3zcg5omva3b6yapxiu

Few-Shot Object Detection: A Survey [article]

Mona Köhler, Markus Eisenbach, Horst-Michael Gross
2021 arXiv   pre-print
To avoid the need to acquire and annotate these huge amounts of data, few-shot object detection aims to learn from few object instances of new categories in the target domain.  ...  As a result, we emphasize common challenges in evaluation and identify the most promising current trends in this emerging field of few-shot object detection.  ...  Li, “Restoring negative information Gap in One-Shot Object Detection,” arXiv preprint arXiv:2011.04267, in few-shot object detection,” in Conference on Neural Information pp. 1–13,  ... 
arXiv:2112.11699v1 fatcat:d6iubz4ui5abvdccjlkm667vay

A Survey of Deep Learning for Low-Shot Object Detection [article]

Qihan Huang, Haofei Zhang, Mengqi Xue, Jie Song, Mingli Song
2022 arXiv   pre-print
Low-Shot Object Detection (LSOD) is an emerging research topic of detecting objects from a few or even no annotated samples, consisting of One-Shot Object Detection (OSOD), Few-Shot Object Detection (FSOD  ...  Although few-shot learning and zero-shot learning have been extensively explored in the field of image classification, it is indispensable to design new methods for object detection in the data-scarce  ...  To achieve this goal, Low-Shot Object Detection (LSOD) is introduced into object detection, including One-Shot Object Detection (OSOD), Few-Shot Object Detection (FSOD), Zero-Shot Object Detection (ZSD  ... 
arXiv:2112.02814v3 fatcat:54s6meub5rcrtiaeo4ylzviofy

Few-Shot Video Object Detection [article]

Qi Fan, Chi-Keung Tang, Yu-Wing Tai
2022 arXiv   pre-print
We introduce Few-Shot Video Object Detection (FSVOD) with three contributions to real-world visual learning challenge in our highly diverse and dynamic world: 1) a large-scale video dataset FSVOD-500 comprising  ...  Extensive experiments demonstrate that our method produces significantly better detection results on two few-shot video object detection datasets compared to image-based methods and other naive video-based  ...  This research was supported in part by Kuaishou Technology, and the Research Grant Council of the Hong Kong SAR under grant No. 16201420.  ... 
arXiv:2104.14805v3 fatcat:pnhjjgvicjcwdj5dt3yw23ha34

Disentangled Feature Representation for Few-shot Image Classification [article]

Hao Cheng, Yufei Wang, Haoliang Li, Alex C. Kot, Bihan Wen
2021 arXiv   pre-print
In general, most of the popular deep few-shot learning methods can be plugged in as the classification branch, thus DFR can boost their performance on various few-shot tasks.  ...  In this work, we propose a novel Disentangled Feature Representation framework, dubbed DFR, for few-shot learning applications.  ...  classification, fine-grained classification, and domain generalization) under the few-shot settings, evaluate the effectiveness of the proposed DFR framework.  ... 
arXiv:2109.12548v1 fatcat:4zsghy7om5h6lkcl6enhxbin5a

Front Matter: Volume 12084

Wolfgang Osten, Dmitry Nikolaev, Jianhong Zhou
2022 Fourteenth International Conference on Machine Vision (ICMV 2021)  
The publisher is not responsible for the validity of the information or for any outcomes resulting from reliance thereon.  ...  of Siamese neural networks in re- identification 0Y Few-shot object detection via metric learning [12084-25] 0Z A novel machine learning approach based on fast multi-scale hybrid wavelet network for supporting  ...  a bounding box [12084-3] 0F Few-shot object detection with anti-confusion grouping [12084-12] 0GParameter decoupling strategy for semi-supervised 3D left atrium segmentationiii Proc. of SPIE Vol. 12084  ... 
doi:10.1117/12.2625908 fatcat:zrgauhqj7ng65flfcmbhqx2ony

Categorical Vehicle Classification and Tracking using Deep Neural Networks

Deependra Sharma, Zainul Abdin Jaffery
2021 International Journal of Advanced Computer Science and Applications  
detections with high precision and boosted efficient binary local image descriptor for tracking multiple vehicle objects are all incorporated into the research.  ...  highway layouts, forcing the system to make trade-offs in terms of the number of actual detections.  ...  Few results of SSMD approach for categorical vehicle detection id depicted in Fig. 8 . C.  ... 
doi:10.14569/ijacsa.2021.0120964 fatcat:k2a6dtf3grepdnflyg3yrjpxzu

Practical Quality Assessment for Digitized Film Content

Francois Helt, Valerie La Torre
2014 SMPTE Motion Imaging Journal  
The Restoration Problem The archivist would like to restore content. He selects few sequenc- es and proposes a few restoration providers to restore them.  ...  There are many variants for the shot boundary detection; this method is based on the structural similarity index combined with histogram differences computation.  ... 
doi:10.5594/j18368xy fatcat:77z2qc5hxrgshb4iw7ujtyhqqi

Automatic Detection of 3D Quality Defects in Stereoscopic Videos Using Binocular Disparity

Sotirios Delis, Ioannis Mademlis, Nikos Nikolaidis, Ioannis Pitas
2017 IEEE transactions on circuits and systems for video technology (Print)  
In this paper, we propose four algorithms that exploit available stereo disparity information, in order to detect disturbing stereoscopic effects, namely stereoscopic window violations (SWV), bent window  ...  effects, UFO objects and depth jump cuts on stereo videos.  ...  The European Union is not liable for any use that may be made of the information contained herein.  ... 
doi:10.1109/tcsvt.2015.2511518 fatcat:g32y4u7hgnc2vbc77epaj7f3ia

DMnet: A New Few-Shot Framework for Wind Turbine Surface Defect Detection

Jinyun Yu, Kaipei Liu, Liang Qin, Qiang Li, Feng Zhao, Qiulin Wang, Haofeng Liu, Boqiang Li, Jing Wang, Kexin Li
2022 Machines  
Experimental results show that the proposed method has significantly improved the defect recognition ability under few-shot training conditions.  ...  In the field of wind turbine surface defect detection, most existing defect detection algorithms have a single solution with poor generalization to the dilemma of insufficient defect samples and have unsatisfactory  ...  Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/machines10060487 fatcat:il36fa5udjhwro3pntbbtd5udi

A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities [article]

Yisheng Song, Ting Wang, Subrota K Mondal, Jyoti Prakash Sahoo
2022 arXiv   pre-print
Despite the recent creative works in tackling FSL tasks, learning valid information rapidly from just a few or even zero samples still remains a serious challenge.  ...  Few-shot learning (FSL) has emerged as an effective learning method and shows great potential.  ...  Few-shot Object Detection Few-Shot Object Detection (FSOD) is the task of detecting rare objects from several samples.  ... 
arXiv:2205.06743v2 fatcat:xmxht2ileja53o2o5b4vrw32ey
« Previous Showing results 1 — 15 out of 19,051 results