https://sgvr.kaist.ac.kr/publication/kinematic-fields/
TopEuropean Conference on Computer Vision (ECCV), 2024
This paper presents a novel approach for reconstructing dynamic radiance fields from monocular videos. We integrate kinematics with dynamic radiance fields, bridging the gap between the sparse nature of monocular … Continue reading "Regularizing Dynamic Radiance Fields with Kinematic Fields"
https://sgvr.kaist.ac.kr/publication/take-a-prior-from-other-tasks-for-severe-blur-removal/
Computer Vision and Image Understanding (CVIU), 2024
Recovering clear structures from severely blurry inputs is a huge challenge due to the detail loss and ambiguous semantics. Although segmentation maps can help deblur facial images, their effectiveness is … Continue reading "Take a Prior from Other Tasks for Severe Blur Removal"
https://sgvr.kaist.ac.kr/publication/uforecon/
TopIEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
https://sgvr.kaist.ac.kr/publication/topological-ransac-for-instance-verification-and-retrieval-without-fine-tuning/
TopThirty-seventh Conference on Neural Information Processing Systems (NeurIPS), 2023
https://sgvr.kaist.ac.kr/publication/ksc2022/
한국소프트웨어종합학술대회 (KSC), 2022
Abstract 본 연구는 행동 예측기를 통해서 시나리오를 생성하는 문제를 탐구한다. 시나리오란 행동의 시퀀스로서, 소셜로봇이 사용자와 협업을 수행하는 데 이용될 수 있다. 하지만 수집된 데이터만으로는 시나리오의 종류가 한정적이기 때문에 다양한 시나리오를 … Continue reading "행동 Scene-Graph 예측기를 활용한 시나리오 생성"
https://sgvr.kaist.ac.kr/publication/contour-aware-equipotential-learning-for-semantic-segmentation/
IEEE Transactions on Multimedia (TMM), 2022
https://sgvr.kaist.ac.kr/publication/icip2022-ada/
IEEE International Conference on Image Processing (ICIP), 2022
https://sgvr.kaist.ac.kr/publication/part-based-pseudo-label-refinement-for-unsupervised-person-re-identification/
TopIEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022
https://sgvr.kaist.ac.kr/publication/deep-video-inpainting-guided-by-audio-visual-self-supervision-2/
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2022
https://sgvr.kaist.ac.kr/publication/bmvc2021-in-n-out/
British Machine Vision Conference (BMVC), 2021
Abstract In computer vision, recovering spatial information by filling in masked regions, e.g., inpainting, has been widely investigated for its usability and wide applicability to other various applications: image inpainting, … Continue reading "In-N-Out: Towards Good Initialization for Inpainting and Outpainting"
https://sgvr.kaist.ac.kr/publication/hypergraph-propagation-and-community-selection-for-objects-retrieval/
Top35th Conference on Neural Information Processing Systems (NeurIPS), 2021
https://sgvr.kaist.ac.kr/publication/unsupsimflow/
TopEuropean Conference on Computer Vision (ECCV), 2020
https://sgvr.kaist.ac.kr/publication/single-image-reflection-removal-with-physically-based-training-images/
TopComputer Vision and Pattern Recognition (CVPR), 2020
Oral paper
https://sgvr.kaist.ac.kr/publication/real-time-3-d-mapping-with-estimating-acoustic-materials/
IEEE/SICE International Symposium on System Integration (SII), 2020
https://sgvr.kaist.ac.kr/publication/an-objectness-score-for-accurate-and-fast-detection-during-navigation/
The World Congress on Advances in Nano, Bio, Robotics and Energy (ANBRE), 2019