Dongyoon Han
Dongyoon Han publishes mostly in Computer Vision and Pattern Recognition, Artificial Intelligence and Mechanical Engineering, on topics such as Domain Adaptation and Few-Shot Learning, Advanced Neural Network Applications and Multimodal Machine Learning Applications.
- World rank
- #925,514 of 7,710,434 ranked researchers
- Rank in South Korea
- #8,808 of 162,609
- Works
- 134
- Citations
- 11k
- Citations per work
- 79.0
What does Dongyoon Han research?
Dongyoon Han's work is mostly in Computer Science (63%), Engineering (19%).
Their six largest subfields, inside their fields and domains; the grey slice is everything else they publish. The interactive view splits every one of their works down to single topics.
The fields and subfields, with their numbers
- Physical Sciences10981.7%
- Computer Science8462.6%
- Engineering2619.1%
- Everything else2518.3%
- 1Computer Vision and Pattern Recognition39% of their works
- 2Artificial Intelligence24% of their works
- 3Mechanical Engineering7% of their works
- 4Electrical and Electronic Engineering5% of their works
- 5Biomedical Engineering4% of their works
- 6Control and Systems Engineering3% of their works
Research topics
- Domain Adaptation and Few-Shot Learning
- Advanced Neural Network Applications
- Multimodal Machine Learning Applications
- Advanced Image and Video Retrieval Techniques
- Adversarial Robustness in Machine Learning
- Generative Adversarial Networks and Image Synthesis
- Anomaly Detection Techniques and Applications
- Handwritten Text Recognition Techniques
Which keywords describe Dongyoon Han's research?
The keywords of their largest research topics: Deep Learning, Neural Networks, Representation Learning, Unsupervised Learning, Convolutional Neural Networks, Domain Adaptation, Transfer Learning and Image Recognition.
- Text Detection
- Anomaly Detection
- Generative Adversarial Networks
- Adversarial Examples
- Feature Matching
- Image Captioning
- Image Recognition
- Domain Adaptation
- Unsupervised Learning
- Neural Networks
- Deep Learning
- Representation Learning
- Convolutional Neural Networks
- Transfer Learning
- Object Detection
- Visual Question Answering
- Local Descriptors
- Robustness
- Image Synthesis
- Handwriting Recognition
- Unsupervised
Size is their works in the topics tagged with each word, from their 8 largest topics. Each links to the topic it comes from most. The words come from the keyword list OpenAlex gives each topic, not from the text of the papers, so a word can appear that none of the papers uses.
All 21 words, with their numbers
- Deep Learning1116 topics
- Neural Networks605 topics
- Representation Learning512 topics
- Unsupervised Learning512 topics
- Convolutional Neural Networks472 topics
- Domain Adaptation421 topic
- Transfer Learning421 topic
- Image Recognition391 topic
- Object Detection391 topic
- Image Captioning231 topic
- Visual Question Answering231 topic
- Feature Matching201 topic
- Local Descriptors201 topic
- Adversarial Examples121 topic
- Robustness121 topic
- Generative Adversarial Networks91 topic
- Image Synthesis91 topic
- Anomaly Detection81 topic
- Handwriting Recognition81 topic
- Text Detection81 topic
- Unsupervised81 topic
Where does Dongyoon Han work?
Dongyoon Han's main affiliation in the publication record is Naver (South Korea), South Korea.
How many publications and citations does Dongyoon Han have?
OpenAlex counts 134 works and 10,582 citations for Dongyoon Han, an h-index of 28, ranking #925,514 of 7,710,434 researchers worldwide on the quality score.
Ranked on field-normalised citation excellence (50%), output (30%) and citations (20%). Counts come from OpenAlex author records, which occasionally merge different people who share a name or split one person into several.
Papers, co-authors, who cited this work and researchers on the nearest topics have tabs of their own, above. Is this your page? Claim it with your ORCID iD, or request a correction or removal.
Discussion
Ask a question or share a note about Dongyoon Han. Keep it to the research: no personal attacks, advertising or links. Any reader can report a post, which collapses it for everyone until it is opened.