(* indicates equal contribution)
Computer vision is key elements for VR, AR and Metaverse. Our research currently addresses the multi-view stereo estimation, Human action control, human pose estimation, and hand gesture recognition.
Semantic Segmentation
Semantic segmentation is one of the most fundamental computer vision tasks, and it is used extensively in many real-world applications, such as autonomous driving and medical diagnoses.
•
Jae-hun Shim*, Hyunwoo Yu*, Kyeongbo Kong*, Suk-Ju Kang, "FeedFormer: Revisiting Transformer Decoder for Efficient Semantic Segmentation", AAAI, 2023. (Top-tier AI conference, Acceptance Rate: 19.6%)
Mosaic-based Omnidirectional Multi-view Stereo for Indoor Scenes
MosaicMVS estimates the depth of the target image using source images from a novel mosaic-based omnidirectional MVS camera setup.
•
Min-jung Shin*, Woojune Park*, Minji Cho*, Kyeongbo Kong*, Hosung Son, Joonsoo Kim, Kugjin Yun, Gwangsoon Lee, Suk-Ju Kang, "MosaicMVS: Mosaic-based Omnidirectional Multi-view Stereo for Indoor Scenes", IEEE Transactions on Multimedia, 2022. (IF: 8.182)
•
Min-jung Shin*, Minji Cho, Woojune Park, Kyeongbo Kong, Joonsoo Kim, Kugjin Yun, Gwangsoon Lee, Suk-Ju Kang, "Mosaic-based omnidirectional depth estimation for view synthesis", ECCV workshop on Learning to Generate 3D Shapes and Scenes (ECCVw), Oct. 2022.
3D Human Action Control
Action-conditioned transformer VAE has shown its ability to generate realistic and diverse human motion sequences. Taking a step further, we want to control the specific body part of the generated human motions, thereby achieving more degrees of freedom and diversity in human actions.
•
Hyunsung Kim*, Kyeongbo Kong*, Joseph Kihoon Kim*, James Lee*, Geonho Cha, Ho-Deok Jang, Dongyoon Wee, Suk-Ju Kang, "3D Human Motion Control in Latent Space of VAE", (Submitted)
Human Pose Estimation
Human pose estimation aims to precisely localize the semantic keypoints of human bodies in an image.
•
Ginam Kim*, Hyunsung Kim*, Kyeongbo Kong, Jou Won Song, Suk-Ju Kang, "Human Body-Aware Feature Extractor Using Attachable Feature Corrector for Human Pose Estimation", IEEE Transactions on Multimedia (TMM), 2022. (IF: 8.182)
Hand Gesture Recognition
Hand gesture recognition is essential to human computer interaction as the most natural way of communicating.
•
Jae-Hun Song, Kyeongbo Kong, Suk-Ju Kang, "Dynamic Hand Gesture Recognition using Improved Spatio-Temporal Graph Convolutional Network", IEEE Transactions on Circuits and Systems for Video Technology, 2022. (IF: 5.859)
Motion Estimation
Motion estimation is the process of determining motion vectors that describe the transformation from one 2D image to another; usually from adjacent frames in a video sequence.
•
Kyeongbo Kong, Seungjun Shin, Woo-Jin Song, "Histogram-based Non-Iterative Global Motion Estimation", ITC-CSCC, 2016. (Oral)
•
Kyeongbo Kong, Seungjun Shin, Junggi Lee, Woo-Jin Song, "How to Estimate Global Motion Non-Iteratively From a Coarsely Sampled Motion Vector Field", IEEE Transactions on Circuits and Systems for Video Technology, 2019. (IF: 4.133)
•
Junggi Lee*, Kyeongbo Kong*, Gyu Jin Bae, Woo-Jin Song, "BlockNet: A Deep Neural Network for Block-Based Motion Estimation Using Representative Matching", SYMMETRY-BASEL, 2020.