Filters








432 Hits in 9.5 sec

Cluster, Split, Fuse, and Update: Meta-Learning for Open Compound Domain Adaptive Semantic Segmentation [article]

Rui Gong, Yuhua Chen, Danda Pani Paudel, Yawei Li, Ajad Chhatkuli, Wen Li, Dengxin Dai, Luc Van Gool
2020 arXiv   pre-print
In this work, we propose a principled meta-learning based approach to OCDA for semantic segmentation, MOCDA, by modeling the unlabeled target domain continuously.  ...  Open compound domain adaptation (OCDA) is a domain adaptation setting, where target domain is modeled as a compound of multiple unknown homogeneous domains, which brings the advantage of improved generalization  ...  MOCDA is composed of four modules, cluster, split, fuse and update module. Meta-learning serves in the fuse and update module for continuously modeling the compound target domain and online update.  ... 
arXiv:2012.08278v1 fatcat:b35synnqyzgozcsbjstspj6epu

Cluster, Split, Fuse, and Update: Meta-Learning for Open Compound Domain Adaptive Semantic Segmentation

Rui Gong, Yuhua Chen, Danda Pani Paudel, Yawei Li, Ajad Chhatkuli, Wen Li, Dengxin Dai, Luc Van Gool
2021
In this work, we propose a principled meta-learning based approach to OCDA for semantic segmentation, MOCDA, by modeling the unlabeled target domain continuously.  ...  Open compound domain adaptation (OCDA) is a domain adaptation setting, where target domain is modeled as a compound of multiple unknown homogeneous domains, which brings the advantage of improved generalization  ...  This research has received funding from the EU Horizon 2020 research and innovation programme under grant agreement No. 820434.  ... 
doi:10.3929/ethz-b-000556709 fatcat:ubfxtjbxffgjviealxrrqevvra

Unsupervised Domain Adaptation for Semantic Image Segmentation: a Comprehensive Survey [article]

Gabriela Csurka, Riccardo Volpi, Boris Chidlovskii
2021 arXiv   pre-print
We present the most important semantic segmentation methods; we provide a comprehensive survey on domain adaptation techniques for semantic segmentation; we unveil newer trends such as multi-domain learning  ...  , domain generalization, test-time adaptation or source-free domain adaptation; we conclude this survey by describing datasets and benchmarks most widely used in semantic segmentation research.  ...  ter, split, fuse, and update: Meta-learning for open com- In IEEE International Conference on Computer Vision pound domain adaptive semantic segmentation.  ... 
arXiv:2112.03241v1 fatcat:uzlehddvuvfwzf4dfbjimja45e

Video Big Data Analytics in the Cloud: A Reference Architecture, Survey, Opportunities, and Open Research Issues

Aftab Alam, Irfan Ullah, Young-Koo Lee
2020 IEEE Access  
Finally, we identify and articulate several open research issues and challenges, which have been raised by the deployment of big data technologies in the cloud for video big data analytics.  ...  The current technology and market trends demand an efficient framework for video big data analytics.  ...  For the positions of key segments and content depiction in the video, they used an adaptive latent structural SVM model and semantic concepts, correspondingly.  ... 
doi:10.1109/access.2020.3017135 fatcat:qc62bhzlrfcwblnvurb5okfjxe

Image retrieval

Ritendra Datta, Dhiraj Joshi, Jia Li, James Z. Wang
2008 ACM Computing Surveys  
While the last decade laid foundation to such promise, it also paved the way for a large number of new techniques and systems, got many new people involved, and triggered stronger association of weakly  ...  We also discuss significant challenges involved in the adaptation of existing image retrieval techniques to build systems that can be useful in the real-world.  ...  Active learning using SVMs were proposed for relevance feedback [Tong and Chang 2001] and helped popularize active learning in other domains as well.  ... 
doi:10.1145/1348246.1348248 fatcat:5jbcrsxkkbac5cya3zb7eb22ea

Domain-Specific Priors and Meta Learning for Few-Shot First-Person Action Recognition [article]

Huseyin Coskun, Zeeshan Zia, Bugra Tekin, Federica Bogo, Nassir Navab, Federico Tombari, Harpreet Sawhney
2021 arXiv   pre-print
We employ a framework based on meta-learning to extract the distinctive and domain invariant components of the deployed visual cues.  ...  We aim to develop an effective method for few-shot transfer learning for first-person action classification.  ...  ACKNOWLEDGMENTS The authors would like to thank David Joseph Tan for the valuable discussions and constructive feedback. This work was supported by Microsoft.  ... 
arXiv:1907.09382v2 fatcat:aj7rdwx5ongd7dd2tsk6ybyeuu

RoboSherlock: Cognition-enabled Robot Perception for Everyday Manipulation Tasks [article]

Ferenc Bálint-Benczédi, Jan-Hendrik Worch and Daniel Nyga and Nico Blodow and Patrick Mania and Zoltán-Csaba Márton and Michael Beetz
2019 arXiv   pre-print
enable automatic and knowledge-driven generation of processing pipelines.  ...  We present RoboSherlock, a knowledge-enabled cognitive perception systems for mobile robots performing human-scale everyday manipulation tasks.  ...  a region for fusing results from multiple images.  ... 
arXiv:1911.10079v1 fatcat:hnude4inmfcs5ncfwgjclkftam

Knowledge Augmented Machine Learning with Applications in Autonomous Driving: A Survey [article]

Julian Wörmann, Daniel Bogdoll, Etienne Bührle, Han Chen, Evaristus Fuh Chuo, Kostadin Cvejoski, Ludger van Elst, Tobias Gleißner, Philip Gottschall, Stefan Griesche, Christian Hellert, Christian Hesels (+34 others)
2022 arXiv   pre-print
The existence of representative datasets is a prerequisite of many successful artificial intelligence and machine learning models.  ...  The reasons for this are manifold and range from time and cost constraints to ethical considerations.  ...  Afterwards the distance of each input datapoint to the cluster is calculated and the closest cluster wins and updates the cluster center.  ... 
arXiv:2205.04712v1 fatcat:u2bgxr2ctnfdjcdbruzrtjwot4

Towards holistic scene understanding: Semantic segmentation and beyond [article]

Panagiotis Meletis
2022 arXiv   pre-print
Chapter 3 focuses on enriching semantic segmentation with weak supervision and proposes a weakly-supervised algorithm for training with bounding box-level and image-level supervision instead of only with  ...  First, we investigate semantic segmentation in the context of street scenes and train semantic segmentation networks on combinations of various datasets.  ...  First, I would like to thank all current and past members of the MPS cluster for spending quality time together and having interesting conversations.  ... 
arXiv:2201.07734v1 fatcat:qdqnjqn75rff7kyja2iwer75my

2021 Index IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Vol. 14

2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Departments and other items may also be covered if they have been judged to have archival value. The Author Index contains the primary entry for each item, listed under the first author's name.  ...  The primary entry includes the coauthors' names, the title of the paper or other item, and its location, specified by the publication abbreviation, year, month, and inclusive pagination.  ...  ., +, JSTARS 2021 9609-9623 STransFuse: Fusing Swin Transformer and Convolutional Neural Network for Remote Sensing Image Semantic Segmentation.  ... 
doi:10.1109/jstars.2022.3143012 fatcat:dnetkulbyvdyne7zxlblmek2qy

Research Contribution and Comprehensive Review towards the Semantic Segmentation of Aerial Images Using Deep Learning Techniques

P. Anilkumar, P. Venugopal, Mamoun Alazab
2022 Security and Communication Networks  
Semantic segmentation is a significant research topic for decades and has been employed in several applications.  ...  In recent years, semantic segmentation has been focused on different deep learning approaches in the area of computer vision, which has aimed for getting superior efficiency while analyzing the aerial  ...  Also, the authors would like to thank the individual copyright holders for consent conceded to incorporate referred figures in this work. 26 Security and Communication Networks  ... 
doi:10.1155/2022/6010912 fatcat:qxoogfb3zneypkh5w3m5p3ts3e

Semantic patch inference

Jesper Andersen, Anh Cuong Nguyen, David Lo, Julia L. Lawall, Siau-Cheng Khoo
2012 Proceedings of the 27th IEEE/ACM International Conference on Automated Software Engineering - ASE 2012  
For example the open-brace token, { can denote both the start of a method body and the start of a compound statement.  ...  For illustration, when fusing the left hand side (non-abstract) patterns we used the meta-variables X0 and X1 while when fusing the right hand side (non-abstract) patterns, we used X1 and X3.  ...  The loop function first fuses the patches in tu _ lists 1 and then combines each of the fused patches with one from tu _ lists i .  ... 
doi:10.1145/2351676.2351753 dblp:conf/kbse/AndersenNLLK12 fatcat:4pnomtho6rbhji47sv6u2jhlt4

Action Recognition in Videos: from Motion Capture Labs to the Web [article]

Ana Paula Brandão Lopes, Eduardo Alves do Valle Jr., Jussara Marques de Almeida, Arnaldo Albuquerque de Araújo
2010 arXiv   pre-print
That perspective is the basis for the discussion in the end of the paper, where we also present the main open issues in the area.  ...  The proposed organization is based on the representation used as input for the recognition task, emphasizing the hypothesis assumed and thus, the constraints imposed on the type of video that each technique  ...  Ivan Laptev for his comments on earlier versions of this manuscript, as well as the Brazilian funding agencies CAPES, CNPq and FAPEMIG.  ... 
arXiv:1006.3506v1 fatcat:rti7aqnwxfgwdivslh2c7w2wly

A Roadmap for Big Model [article]

Sha Yuan, Hanyu Zhao, Shuai Zhao, Jiahong Leng, Yangxiao Liang, Xiaozhi Wang, Jifan Yu, Xin Lv, Zhou Shao, Jiaao He, Yankai Lin, Xu Han (+88 others)
2022 arXiv   pre-print
With the rapid development of deep learning, training Big Models (BMs) for multiple downstream tasks becomes a popular paradigm.  ...  In this paper, we cover not only the BM technologies themselves but also the prerequisites for BM training and applications with BMs, dividing the BM review into four parts: Resource, Models, Key Technologies  ...  and semantic segmentation etc.  ... 
arXiv:2203.14101v4 fatcat:rdikzudoezak5b36cf6hhne5u4

A review of uncertainty quantification in deep learning: Techniques, applications and challenges

Moloud Abdar, Farhad Pourpanah, Sadiq Hussain, Dana Rezazadegan, Li Liu, Mohammad Ghavamzadeh, Paul Fieguth, Xiaochun Cao, Abbas Khosravi, U. Rajendra Acharya, Vladimir Makarenkov, Saeid Nahavandi
2021 Information Fusion  
This study reviews recent advances in UQ methods used in deep learning, investigates the application of these methods in reinforcement learning, and highlights fundamental research challenges and directions  ...  (e.g., image restoration), medical image analysis (e.g., medical image classification and segmentation), natural language processing (e.g., text classification, social media texts and recidivism risk-scoring  ...  Acknowledgment This work was partially supported by the Australian Research Council's Discovery Projects funding scheme (project DP190102181) and the Natural Sciences and Engineering Research Council of  ... 
doi:10.1016/j.inffus.2021.05.008 fatcat:yschhguyxbfntftj6jv4dgywxm
« Previous Showing results 1 — 15 out of 432 results