The Deep Computer Vision Laboratory (DCVL) is directed by professor Wonjun Kim since 2016. Research in our lab focuses on understanding a given image and video in a computational way. We aspire to build up intelligent methods that perform innovative visual tasks such as scene understanding, image enhancement/restoration, vision-based biometrics, etc.
We are looking for students who want to join our lab ! If you are interested in computer vision, image processing, and machine learning, feel free to contact me : wonjkim(at)konkuk.ac.kr (more info)
Jingi Kim and Wonjun Kim, "SVD-guided diffusion for training-free low-light image enhancement," IEEE Signal Processing Letters, accepted, Aug. 2025.
Gun Ryu and Wonjun Kim, "Shape-selective splatting: Regularizing the shape of Gaussian for sparse-view rendering," IEEE Signal Processing Letters, accepted, Jul. 2025.
SungMin Jang and Wonjun Kim, "Identity-aware language Gaussian splatting for open-vocabulary 3D semantic segmentation," in Proc. IEEE/CVF International Conference on Computer Vision (ICCV), Oct. 2025.
Mi-Gyeong Gwon, G.-M. Um, W. S. Cheong, and Wonjun Kim, "Eigenpose: Occlusion-robust 3D human mesh reconstruction," IEEE Transactions on Image Processing (IF=13.7, JCR top 2%), accepted, Mar. 2025.
Hyunwoo Park, Gun Ryu, and Wonjun Kim, "DropGaussian: Structural regularization for sparse-view Gaussian splatting," in Proc. IEEE/CVF International Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2025.
Our papers are accepted for IEEE Signal Processing Letters (Jul. 2025, Aug. 2025)
Our paper is accepted for IEEE/CVF Conference on Computer Vision (ICCV) (Oct. 2025).
Our paper is accepted for IEEE Signal Processing Letters (Apr. 2025)
Our paper is accepted for IEEE Transactions on Image Processing (Mar. 2025)
Prof. Wonjun Kim serves as Program Chair of IPIU 2025 (Feb. 2025)
Our paper is accepted for IEEE Access (Dec. 2023)
Our paper is accepted for Multimedia Tools and Applications (Nov. 2023)
Our paper is accepted for Multimedia Systems (Nov. 2023)
Our paper is accepted for Journal of Visual Communication and Image Representation (Jun. 2023)
Prof. Wonjun Kim serves as VSPC-TC members (IEEE Circuits and Systems Society)
Our paper is accepted for IEEE Access (Oct. 2022)
Our paper is accepted for IEEE Transactions on Multimedia (Oct. 2022)
Our paper is accepted for IEEE Access (Aug. 2022)
Our paper is accepted for Journal of Visual Communication and Image Representation (Apr. 2022)
Our paper is accepted for Multimedia Tools and Applications (Feb. 2022)
Our paper is accepted for Journal of Visual Communication and Image Representation (Oct. 2021)
권미경(학부연구생), 송민수(석사) 학생이 대한전자공학회 하계학술대회에서 네이버 관심 논문상을 수상 (Jul. 1, 2021)
Our paper is accepted for IEEE Transactions on Circuits and Systems for Video Technology (Jan. 2021)
Our paper is accepted for IEEE Transactions on Multimedia (Nov. 2020)
[Code] Our project website for SVD-guided diffusion for low-light image enhancement (SPL) now opens in github : https://github.com/DCVL-LLIE/SGD-release
[Code] Our project website for Shape-selective splatting (SPL) now opens in github : https://github.com/DCVL-3D/SSS_release
[Code] Our project website for identity-aware language Gaussian splatting (ICCV 2025) now opens in github : https://github.com/DCVL-3D/ILGS_release
[Code] Our project website for Eigenpose (TIP) now opens in github : https://github.com/DCVL-3D/Eigenpose_release
[Code] Our project website for clustering-based adaptive query generation for semantic segmentation (SPL) now opens in github : https://github.com/DCVL-Seg/CQG-Release
[Code] Our project website for DropGaussian (CVPR 2025) now opens in github : https://github.com/DCVL-3D/DropGaussian_release
[Code] Our project website for Occlusion-robust 3D human mesh reconstruction (CVPR 2024) now opens in github : https://github.com/DCVL-3D/InstanceHMR_release
[Code] Our project website for weakly-supervised semantic segmenation now opens in github : https://github.com/DCVL-WSSS/ClusterCAM
[Code] Our project website for instance segmentation (TMM) now opens in github : https://github.com/tjqansthd/LapMask
[Code] Our project website for 3D human mesh reconstruction (CVPR 2023) now opens in github : https://github.com/DCVL-3D/PointHMR_release
[Code] Our project website for knowledge distillation via Laplacian decomposition (JVCI) now opens in github : https://github.com/tjqansthd/Lap_Rep_KD_Depth
[Code] Our project website for Laplacian decomposition-based depth estimation (TCSVT) now opens in github : https://github.com/tjqansthd/LapDepth-release
[Code] Our project website for low-light image enhancement (TMM) now opens in github : https://github.com/SeokjaeLIM/DSLR-release
[Code] Our project website for shadow detection (SPL) now opens in github : https://github.com/JinheeKIM94/AFFPN_release
[Code] Our project website for image deblurring (SPL) now opens in github : https://github.com/SeokjaeLIM/DSSN_release
[Code] Our project website for face anti-spoofing using stereo images now opens in github : https://github.com/dcvl18/DeepDualNet-Pytorch
[Code] Our project website for depth estimation from a single image now opens in github : https://github.com/tjqansthd/GDN-Pytorch