중앙대학교 첨단영상대학원

구성원

전임

Weakly-/Semi-supervised Perception, 3D Vision, Domain Adaptation/Generalization, AI Personalization, Human-AI Collaboration

저희 연구실은 개개인이 대규모 데이터나 고성능 연산 자원 없이도 비전 AI를 개발하고 활용할 수 있는 미래를 목표로 합니다. Weakly-/Semi-supervised Perception, 3D Vision, Domain Adaptation/Generalization, AI Personalization, Human-AI Collaboration 등의 기술들을 연구합니다.

2024 - Qualcomm Innovation Fellowship 2024 Korea, Winner
2023 - 29th Samsung HumanTech Paper Awards, Bronze prize
2023 - 35th Workshop on Image Processing and Image Understanding (IPIU 2023), Bronze prize in Best Paper Awards / Best poster presentation prize
2022 - 34th Workshop on Image Processing and Image Understanding (IPIU 2023), Gold prize in Best Paper Awards

From SAM to CAMs: Exploring Segment Anything Model for Weakly Supervised Semantic Segmentation, CVPR 2024, Selected as oral presentation (Top 3.3% of accepted)
WISH: Weakly Supervised Instance Segmentation using Heterogeneous Labels, CVPR 2025
TALoS: Enhancing Semantic Scene Completion via Test-time Adaptation on the Line of Sight, NeurIPS 2024
Weakly Supervised Point Cloud Semantic Segmentation via Artificial Oracle, CVPR 2024
Weakly Supervised Semantic Segmentation via Adversarial Learning of Classifier and Reconstructor, CVPR 2023
Joint Learning of 2D-3D Weakly Supervised Semantic Segmentation, NeurIPS 2022