Filters








470 Hits in 6.0 sec

Bongard-LOGO: A New Benchmark for Human-Level Concept Learning and Reasoning [article]

Weili Nie, Zhiding Yu, Lei Mao, Ankit B. Patel, Yuke Zhu, Animashree Anandkumar
2021 arXiv   pre-print
Even though today's machine learning models excel with a plethora of training data on standard recognition tasks, a considerable gap exists between machine-level pattern recognition and human-level concept  ...  Inspired by the original one hundred BPs, we propose a new benchmark Bongard-LOGO for human-level concept learning and reasoning.  ...  We also thank all the human subjects for participating in our BONGARD-LOGO human study, and the entire AIALGO team at NVIDIA for their valuable feedback.  ... 
arXiv:2010.00763v4 fatcat:f2pbxqz27rctthvxr2pvzhre54

DeepLogo: Hitting Logo Recognition with the Deep Neural Network Hammer [article]

Forrest N. Iandola, Anting Shen, Peter Gao, Kurt Keutzer
2015 arXiv   pre-print
We propose several DCNN architectures, with which we surpass published state-of-art accuracy on a popular logo recognition dataset.  ...  Meanwhile, deep convolutional neural networks (DCNNs) have revolutionized a broad range of object recognition applications. In this work, we apply DCNNs to logo recognition.  ...  All of our experiments were conducted with the Caffe [10] deep learning framework.  ... 
arXiv:1510.02131v1 fatcat:d7yrd7kwovfvhdmqmeuziuzvce

Scalable Object Detection for Stylized Objects [article]

Aayush Garg, Thilo Will, William Darling, Willi Richert, Clemens Marschner
2017 arXiv   pre-print
such as logo recognition.  ...  Our first layer is a CNN from the Single Shot Multibox Detector family of models that learns to propose regions where some stylized object is likely to appear.  ...  full "object class -no object class" binary object detection network (a "logo -no logo" model in our running example), and employing a deep, scalable image retrieval system that works with object prototypes  ... 
arXiv:1711.09822v2 fatcat:elmm22osfnes7ck6lyh3vg4poi

Deep Learning Logo Detection with Data Expansion by Synthesising Context [article]

Hang Su, Xiatian Zhu, Shaogang Gong
2018 arXiv   pre-print
In this work, we describe a model training image synthesising method capable of improving significantly logo detection performance when only a handful of (e.g., 10) labelled training images captured in  ...  For benchmarking model performance, we introduce a new logo detection dataset TopLogo-10 collected from top 10 most popular clothing/wearable brandname logos captured in rich visual context.  ...  In model learning, we exploit a sequential learning strategy by first deploying a large number of synthesised images to pre-train a deep model, followed by fine-tuning the deep model with the sparse manually  ... 
arXiv:1612.09322v3 fatcat:rgxmnebndjeatdpmldc5jclwbm

Deep Learning Logo Detection with Data Expansion by Synthesising Context

Hang Su, Xiatian Zhu, Shaogang Gong
2017 2017 IEEE Winter Conference on Applications of Computer Vision (WACV)  
In this work, we describe a model training image synthesising method capable of improving significantly logo detection performance when only a handful of (e.g., 10) labelled training images captured in  ...  For benchmarking model performance, we introduce a new logo detection dataset TopLogo-10 collected from top 10 most popular clothing/wearable brandname logos captured in rich visual context.  ...  In model learning, we exploit a sequential learning strategy by first deploying a large number of synthesised images to pre-train a deep model, followed by fine-tuning the deep model with the sparse manually  ... 
doi:10.1109/wacv.2017.65 dblp:conf/wacv/SuZG17 fatcat:5atqvxatybct7huorwe4doxpym

Opportunities and Challenges in Democratizing Immunology Datasets

Sanchita Bhattacharya, Zicheng Hu, Atul J. Butte
2021 Frontiers in Immunology  
We present use cases for repurposing open-access immunology datasets with advanced machine learning applications and more.  ...  Concurrently, the application of computational methods, such as network analysis, meta-analysis, and machine learning have propelled the field forward by providing insight into salient features that influence  ...  The graphics for "Measurement Techniques" in Figure 1 were created with BioRender.com. The logo credits in Figure 1 are provided in the Supplemental Materials.  ... 
doi:10.3389/fimmu.2021.647536 pmid:33936065 pmcid:PMC8086961 fatcat:x5yp4fhgzfdabjmlqjr6spzuly

A comprehensive survey of AI-enabled phishing attacks detection techniques

Abdul Basit, Maham Zafar, Xuan Liu, Abdul Rehman Javed, Zunera Jalil, Kashif Kifayat
2020 Telecommunications Systems  
To provide a thorough understanding of phishing attack(s), this paper provides a literature review of Artificial Intelligence (AI) techniques: Machine Learning, Deep Learning, Hybrid Learning, and Scenario-based  ...  In recent times, a phishing attack has become one of the most prominent attacks faced by internet users, governments, and service-providing organizations.  ...  Furthermore, the proposed models beat existing ML-based models in phishing attack recognition.  ... 
doi:10.1007/s11235-020-00733-2 pmid:33110340 pmcid:PMC7581503 fatcat:c2chplwvvncmlmiyxjgb7iudam

Skeleton-based Hand-Gesture Recognition with Lightweight Graph Convolutional Networks [article]

Hichem Sahbi
2021 arXiv   pre-print
Graph convolutional networks (GCNs) aim at extending deep learning to arbitrary irregular domains, namely graphs.  ...  In this paper, we introduce a novel method that learns the topology (or connectivity) of input graphs as a part of GCN design.  ...  For instance, [75] proposes a graph network for semi-supervised classification that learns graph topology with sparse structure given a cloud of points; node-to-node connections are modeled with a joint  ... 
arXiv:2104.04255v2 fatcat:pu7elcvjgjfznfgnpsjauzimtm

The Kairntech Sherpa - An ML Platform and API for the Enrichment of (not only) Scientific Content

Stefan Geißler
2020 International Conference on Language Resources and Evaluation  
Dedicated specific packages for subtasks such as document structure processing, document categorization, annotation with existing thesauri, disambiguation and linking, annotation with newly created entity  ...  recognizers and summarizationavailable as open source components in isolationare combined into an end-user-facing, collaborative, scalable platform to support large-scale industrial document analysis.  ...  run with the more resource-intensive Deep Learning libraries.  ... 
dblp:conf/lrec/Geissler20 fatcat:shr4mzpzv5gtxadvw4ud7tgsti

Exploiting Multimedia in Creating and Analysing Multimedia Web Archives

Jonathon Hare, David Dupplaw, Paul Lewis, Wendy Hall, Kirk Martinez
2014 Future Internet  
Web archiving provides a way of capturing snapshots of (parts of) the web for preservation and future analysis.  ...  However, due to the dynamic and distributed nature of the web, its content changes, appears and disappears on a daily basis.  ...  logo to be changed in a bootleg copy on YouTube).  ... 
doi:10.3390/fi6020242 fatcat:hnusw44irfe2nf5v6jvgqta2t4

An Empirical Study of Person Re-Identification with Attributes

Vikram Shree, Wei-Lun Chao, Mark Campbell
2019 2019 28th IEEE International Conference on Robot and Human Interactive Communication (RO-MAN)  
There is, however, a plethora of real-life scenarios where we may not have a priori library of query images and therefore must rely on information from other modalities.  ...  A key conclusion is that the performance achieved by non-expert attributes, instead of expert-annotated ones, is a more faithful indicator of the status quo of attribute-based approaches for person re-identification  ...  With scarce attribute data, learning a lower dimensional mapping from the attribute space to the deep-feature space is easier than learning a high dimensional mapping.  ... 
doi:10.1109/ro-man46459.2019.8956459 dblp:conf/ro-man/ShreeCC19 fatcat:hvy35jssk5fpjp2brzveqkbooq

2021 Index IEEE Transactions on Intelligent Transportation Systems Vol. 22

2021 IEEE transactions on intelligent transportation systems (Print)  
Yin, H., +, TITS Feb. 2021 837-852 A Cascaded Deep Convolutional Network for Vehicle Logo Recognition From Frontal and Rear Images of Vehicles.  ...  Logo Recognition From Frontal and Rear Images of Vehicles.  ...  + Check author entry for coauthors Hisham, A., Strom, E.G., and Brannstrom, F., Radio Li, L., Yao, W., Zhao, Y., Yi, C., Lin, J., and Tsui, K.L., Correction to "High-Speed Rail Suspension System Health  ... 
doi:10.1109/tits.2021.3139738 fatcat:p2mkawtrsbaepj4zk24xhyl2oa

Deep-4mCGP: A Deep Learning Approach to Predict 4mC Sites in Geobacter pickeringii by Using Correlation-Based Feature Selection Technique

Hasan Zulfiqar, Qin-Lai Huang, Hao Lv, Zi-Jie Sun, Fu-Ying Dao, Hao Lin
2022 International Journal of Molecular Sciences  
The purpose of this study was to establish a robust deep learning model to recognize 4mC sites in Geobacter pickeringii.  ...  The performance of the anticipated model on independent data exhibited an accuracy of 0.868, which was 4.2% higher than the existing model.  ...  In this work, a deep learning model was constructed to recognize 4mC sites in Geobacter pickeringii.  ... 
doi:10.3390/ijms23031251 pmid:35163174 pmcid:PMC8836036 fatcat:dwmguupra5e7hlrwmgxmtwbpzy

Recent Advance in Content-based Image Retrieval: A Literature Survey [article]

Wengang Zhou, Houqiang Li, Qi Tian
2017 arXiv   pre-print
With the ignorance of visual content as a ranking clue, methods with text search techniques for visual retrieval may suffer inconsistency between the text words and visual content.  ...  We conclude with several promising directions for future research.  ...  In the above approaches, the learning-based feature is extracted with the deep learning model trained for classification task.  ... 
arXiv:1706.06064v2 fatcat:m52xwsw5pzfzdbxo5o6dye2gde

Semantic Indexing of Wearable Camera Images

Alan F. Smeaton, Louise Signal, Moira Smith, James Stanley, Michelle Barr, Tim Chambers, Cliona Ní Mhurchu, Kevin McGuinness, Cathal Gurrin, Jiang Zhou, Noel E. O'Connor, Peng Wang (+3 others)
2016 Proceedings of the 2016 ACM workshop on Vision and Language Integration Meets Multimedia Fusion - iV&L-MM '16  
While great progress has been made in very recent years in automatic concept detection using machine learning, we are still left with a mis-match between the semantics of the concepts we can automatically  ...  annotated from a vocabulary of 1,000 concepts.  ...  The processing pipeline was implemented in Python using the MXNet deep learning library [3] . 9 .  ... 
doi:10.1145/2983563.2983566 dblp:conf/mm/SmeatonMGZOWDAF16 fatcat:srn2eh46m5bnvcm4563ypt2ogy
« Previous Showing results 1 — 15 out of 470 results