KCCV 2016

2016. 7. 21-22, Ewha Womans University, Seoul

프로그램

Day1 : 2016. 7. 21 (Thr)

시간 발표자 발표제목
13:00 - 13:20 Opening
Special Talks (Chair: 이경무 교수)
13:20 - 14:20 Prof. Jiaya Jia
CUHK
Semantic Segmentation and Object Detection from Images with Two Deep Configurations [slides]
14:20 - 15:20 Prof. Ian Reid
University of Adelaide
Deep SLAM? [slides]
15:20 - 15:50 Break
15:50 - 16:50 Prof. Sato Yoichi
U. of Tokyo
Understanding Group Activities from Multiple Egocentric Videos [slides]
Session 1 (Chair: 이상욱 교수)
16:50 - 17:20 김태균 교수
Imperial College
Pose Estimation Demonstrations and Combined of Decision Forests/CNNs [slides]
17:20 - 17:50 한보형 교수
POSTECH
Visual Question Answering using Convolutional Neural Network with Dynamic Parameter Prediction [slides]

Day2 : 2016. 7. 22 (Fri)

시간 발표자 발표제목
Session 2 (Chair: 임종우 교수)
09:30 - 10:00 김창수 교수
고려대
Image and Video Segmentation [slides]
10:00 - 10:30 윤국진 교수
GIST
Tracking and Identifying Multiple Objects across Multiple Cameras [slides]
10:30 - 11:00 이광무 박사
EPFL
Learning Local Features Through Deep Learning [slides]
11:00 - 11:30 한국 컴퓨터비전 학회 창립총회
11:30 - 13:00 Poster Session 1 & Lunch
김창수 (고려대) Primary Object Segmentation in Videos via Alternate Convex Optimization of Foreground and Background Distributions
윤국진 (GIST) Improving Person Re-identification via Pose-aware Multi-shot Matching
윤국진 (GIST) Online Multi-object Tracking via Structural Constraint Event Aggregation
이광무 (EPFL) Learning Good Orientations to Match
이승용 (포항공대) Texture Map Generation for Large-scale 3D Reconstructed Scenes
최진영 (서울대) Dynamic Formulation of Maximum Weighted Clique Problem for Multiple Camera Multiple Target Tracking
Session 3 (Chair: 박종일 교수)
13:00 - 13:30 김선주 교수
연세대
A Holistic Approach to Cross-Channel Image Noise Modeling & Hyperspectral Imaging with Everyday Digital Cameras [slides]
13:30 - 14:00 박인규 교수
인하대
Depth map estimation from unusual cameras [slides]
14:00 - 14:30 김준모 교수
KAIST
Joint Fine-Tuning in Deep Neural Networks for Facial Expression Recognition [slides]
14:30 - 15:10 Poster Session 2 & Coffee Break
김선주 (연세대) Do It Yourself Hyperspectral Imaging with Everyday Digital Cameras
김선주 (연세대) A Holistic Approach to Cross-Channel Image Noise Modeling and its Application to Image Denoising
김준모 (KAIST) Deep Saliency with Encoded Low level Distance Map and High Level Features
박인규 (인하대) Depth map estimation and colorization of anaglyph images
박인규 (인하대) Robust light field depth estimation for noisy scene with occlusion
황성주 (UNIST) Asymmetric Multi-task Learning based on Task Relatedness and Loss
Session 4 (Chair: 이승용 교수)
15:10 - 15:40 임재환 교수
USC
Visual Understanding of the Physical World for Interaction [slides]
15:40 - 16:10 김민혁 교수
KAIST
Advanced Patch-Based Image Synthesis [slides]
16:10 - 16:40 권인소 교수
KAIST
Robustness in Computer Vision for Practical Applications: from 3D to Recognition
16:40 - 17:20 Poster Session 3 & Coffee Break
김민혁 (KAIST) Laplacian Patch-Based Image Synthesis
김민혁 (KAIST) Multiview Image Completion with Space Structure Propagation
김태균 (Imp. Col.) Hierarchical Sampling Optimization for Estimating Human Hand Pose
김태균 (Imp. Col.) Conditional Convolutional Neural Network for Modality-ware Face Analysis
이경무 (서울대) Discrete Tabu Search for Graph Matching
한보형 (포항공대) Learning Transferrable Knowledge for Semantic Segmentation with Deep Convolutional Neural Network
17:20 - 17:30 Closing


발표자료 : Download (152MB)
All slides are password protected.