Query-Dependent Video Representation for Moment Retrieval and Highlight Detection (CVPR23)
WonJun Moon*1 SangEek Hyun*1 SangUk Park2 Dongchan Park2 Jae-Pil Heo1 Sungkyunkwan University1 Pyler2
About
Hello. I’m WonJun Moon, a Ph.D. student at Sungkyunkwan university, South Korea. I am a member of Visual Computing LABoratory advised by Prof. Jae-Pil Heo. My current research interests primarily lie in Video Understanding and Multi-modal Learning. Specifically, I am mostly interested in Temporal Grounding (Moment Retrieval), Video Retrieval, and Object Segmentation. Besides, my other interests are Representation Learning, Image/Video Segmentation, Object Detection, Open Set Recognition, Semi-Supervised Learning, Few-Shot Learning, and Long-Tailed Recognition.
Publications
(* : equal contribution)
International
| ECCV 2024 | Progressive Proxy Anchor Propagation for Unsupervised Semantic Segmentation
[Arxiv] [Code] |
Hyun Seok Seong, WonJun Moon, SuBeen Lee, Jae-Pil Heo |
|---|---|---|
| ECCV 2024 | Mitigating Background Shift in Class-Incremental Semantic Segmentation
[Arxiv] [Code] |
Gilhan Park, WonJun Moon, SuBeen Lee, Tae-Young Kim, Jae-Pil Heo |
| arXiv 2024 | Correlation-Guided Query-Dependency Calibration for Video Temporal Grounding
[Arxiv] [Code] |
WonJun Moon, Sangeek Hyun, SuBeen Lee, Jae-Pil Heo |
| AAAI 2024 | VLCounter: Text-aware Visual Representation for Zero-Shot Object Counting
[Arxiv] [Paper] [Code] |
Seunggu Kang, WonJun Moon, Euiyeon Kim, Jae-Pil Heo |
| CVPR 2023 | Query-Dependent Video Representation for Moment Retrieval and Highlight Detection
[Arxiv] [Paper] [Code] [Video] |
WonJun Moon*, Sangeek Hyun*, Sanguk Park, Dongchan Park, Jae-Pil Heo |
| CVPR 2023 | Leveraging Hidden Positives for Unsupervised Semantic Segmentation
[Arxiv] [Paper] [Code] |
Hyun Seok Seong, WonJun Moon, SuBeen Lee, Jae-Pil Heo |
| AAAI 2023 (oral) | Minority-Oriented Vicinity Expansion with Attentive Aggregation for Video Long-Tailed Recognition
[Arxiv] [Paper] [Code] [Video] |
WonJun Moon, Hyun Seok Seong, Jae-Pil Heo |
| ECCV 2022 | Tailoring Self-Supervision for Supervised Learning
[Arxiv] [Paper] [Code] [Video] |
WonJun Moon, Ji-Hwan Kim, Jae-Pil Heo |
| ECCV 2022 | Difficulty-Aware Simulator for Open Set Recognition
[Arxiv] [Paper] [Code] [Video] |
WonJun Moon, Junho Park, Hyun Seok Seong, Cheol-Ho Cho, Jae-Pil Heo |
| CVPR 2022 (oral) | Task Discrepancy Maximization for Fine-grained Few-Shot Classification
[Arxiv] [Paper] [Code] |
SuBeen Lee, WonJun Moon, Jae-Pil Heo |
Domestic
| KIISE 2022 | Learning from Data Imbalance with Class Grouping Loss | WonJun Moon, Jae-Pil Heo |
|---|---|---|
| KIISE 2020 | Mix-Contrastive Match | WonJun Moon, Jae-Pil Heo |
Experience & Achievements
- Research Intern in Naver Cloud, hosted by Minho Shim (Oct. 2023 - Mar. 2024)
- 2023 President’s List, Sungkyunkwan University
- 18th Scholarship student of the Kwanjeong Educational Foundation
Reviewer
- CVPR’22,23,24
- ICCV’23
- ECCV’24
Education
-
Ph.D. Dept of Artificial Intelligence, Sungkyunkwan University
-
MS. Dept of Artificial Intelligence, Sungkyunkwan University
GPA : 4.38 / 4.5 -
Bs. Dept of Software, Sungkyunkwan University
GPA : 4.42 / 4.5 -
Language : Korean, English
Recent posts
Minority-Oriented Vicinity Expansion with Attentive Aggregation for Video Long-Tailed Recognition (AAAI23)
WonJun Moon Hyun Seok Seong Jae-Pil Heo Sungkyunkwan University
Tailoring Self-Supervision for Supervised Learning (ECCV22)
Abstract Recently, it is shown that deploying a proper self-supervision is a prospective way to enhance the performance of supervised learning. Yet, the ben...
Difficulty-Aware Simulator for Open Set Recognition (ECCV22)
Abstract Open set recognition (OSR) assumes unknown instances appear out of the blue at the inference time. The main challenge of OSR is that the response o...