Notice

Heountaek Lim 's paper has been accepted to 2018 ICASSP. The title is " VR IQA NET: Deep Virtual Reality Image Quality Assessment using Adversarial Learning ". The paper contribution is to propose a new no-reference VR IQA method that exploits the reference image by deep adversarial learning. This paper has been written by Heountaek Lim, Hak Gu Kim and Yong Man Ro.

Sangmin Lee's paper has been accepted to IEEE International Conference of Acoustics, Speech and Signal Processing 2018. The title is "STAN: Spatio-temporal Adversarial Networks for Abnormal Event Detection". The paper contribution is to propose a novel generative model based abnormal event detection method. This paper has been written by Sangmin Lee, Hak Gu Kim, and Yong Man Ro.

The paper written by Seong Tae Kim, Hakmin Lee, Hak Gu Kim, and Yong Man Ro has been awarded Robert F. Wagner All-Conference Best Student Paper Award Finalist (Best Student Paper Award in Computer-Aided Diagnosis Conference) in SPIE Medical Imaging 2018 held in United States. The awarded paper title is “ICADx: Interpretable computer aided diagnosis of breast masses”, which is outcome of the Explainable AI project.

Jung Uk Kim, a master's in IVY lab under the supervision of Prof. Yong Man Ro, received the best paper award of the Korean Multimedia Society conference which was held in Nov, 2017. The title of the paper is "Face Detection Network Design with Less Parameter and Robust to Wild Environment using Teacher-Student Learning Method." Sincerely congratulations!!

Hak Gu Kim's paper has been accepted to Optics Express. The title is " Ultrafast layer based computer-generated hologram calculation with sparse template holographic fringe pattern for 3-D object ". The paper contribution is to propose a new ultrafast CGH calculation that exploits the sparsity of hologram fringe pattern in 3-D object layer. This paper has been written by Hak Gu Kim and Yong Man Ro.

Wissam's paper has been accepted to 32th AAAI Conference od Artificial Intelligence (AAAI-18). The title is "Learning Spatio-temporal Features woth Partial Expression Sequences for on-the-Fly Prediction". The paper contribution is to propose novel prediction algorithm performed on-the-Fly, which is very useful in predicting facial expression on-the-Fly as they are fed to the system. This paper has been written by Wisam J. Baddar and Yong Man Ro.

Tae Kwan Lee's paper has been accepted to 24th International conference on Multimedia Modeling 2018. The title is "Convolution with Logarithmic Filter Groups for Efficient Shallow CNN". The paper contribution is to propose a novel nonlinear filter grouping method for CNN parameter reduction. This paper has been written by Tae Kwan Lee, Wissam J. Baddar, Seong Tae Kim, and Yong Man Ro.

Hongjoo Lee's paper has been accepted to 24th International conference on Multimedia Modeling 2018. The title is "Teacher and Student Joint Learning for Compact Facial Landmark Detection Network". The paper contribution is to propse novel teacher and student Learning for very compact network, which is very useful in mobile application. This paper has been written by Hong Joo Lee, Wisam J.Baddar, Hak Gu Kim, Seong Tae Kim and Yong Man Ro.

Seong Tae Kim and Hakmin Lee's paper has been accepted to Computer-Aided Diagnosis" conference in SPIE Medical Imaging 2017. The title is "ICADx: interpretable computer aided diagnosis of breast masses". The paper contribution is firstly to provide interpretability to CADx. This paper has been written by Sung Tae Kim, Hakmin Lee, Hak Gu Kim, and Yong Man Ro.

Hak Gu Kim's paper has been accepted to ACM VRST 2017. The title is "Measurement of exceptional motion in VR video contents for VR sickness assessment using deep convolutional autoencoder ". The paper contribution is to devise new and first VR quality (VR sickness) assessment by using deep convolutional autoencoder. This paper has been written by Hak Gu Kim, Wisam J. Baddar, Hyunwook Jeong, Heountaek Lim and Yong Man Ro.

Seong Tae Kim's paper has been accepted to IEEE BIOSIG 2017. The title is "Multi-scale facial scanning via spatial LSTM for latent facial feature representation". The paper contribution is to new latent facial feature representation by spatial LSTM. This paper has been written by Seong Tae Kim, Yeoreum Choi and Yong Man Ro. 

2018년도 전기 국비 박사과정, 석사과정, 산학장학생 (KEPSI, EPSS, LGenius) 등을 모집합니다. (http://admission.kaist.ac.kr/graduate/ )

모집 연구분야: Deep learning, Machine learning in computer vision and image processing, Image processing (2D, 3D, VR), Computer vision, Medical imaging, Visual recognition, Quality Assessment

현재 진행중인 연구과제: Explainable (Interpretable) Deep learning, Deep learning algorithm in image processing/ computer vision, Recognition/Emotion recognition, 3D/VR quality assessment with deep learning approach, Medical Image analysis with deep learning

최근 연구실 연구결과

http://ivylab.kaist.ac.kr/htm/research/project.asp?MainDiv=3)

최근 연구원 딥러닝 관련 해외 학회 발표실적

(http://ivylab.kaist.ac.kr/htm/publication/publication.asp?MainDiv=5)을 참고하세요.

최근 연구원 해외 저널 실적

(http://ivylab.kaist.ac.kr/htm/publication/publication.asp?MainDiv=4)

연구실 들어오고자 하는 학생은 노용만 교수님(ymro@kaist.ac.kr) 께 이메일/사전미팅 추천합니다.

Hak Gu Kim's paper has been accepted to CISP-BMEI 2017. The title is "Modality-bridge transfer learning for medical image classification". The paper contribution is the new transfer learning deployable in small size of medical training images. This paper has been written by Hak Gu Kim and Yong Man Ro. 

Our proud student (M.S candidate) of IVY Lab, Jung Uk Kim's paper has been accepted to the thematic workshops of ACM Multimedia 2017. The paper title is "Robust and Real-Time Visual Tracking with Triplet Convolutional Neural Network". The contribution of paper is to devise new learning algorithm to achieve real time and robust object tracking. This paper has been written by Jung Uk Kim, Hak Gu Kim and Yong Man Ro. 

https://youtu.be/FBBrK5ZgQUs

Hyunwook Jeong's paper has been accepted to IEEE International Conference on Image Processing (ICIP 2017),

which is entitled by “Visual Comfort Assessment of Stereoscopic Images using Deep Visual and Disparity Features Based on Human Attention".

This paper was written by Hyunwook Jeong, Hak Gu Kim, and Yong Man Ro.

Congratulations!!

Jung Uk Kim's paper has been accepted to IEEE Engineering in Medicine & Biology Society (EMBC'17),

which is entitled by “Iterative Deep Convolutional Encoder-Decoder Network for Medical Image Segmentation ".

This paper was written by Jung Uk Kim, Hak Gu Kim, Yong Man Ro.

Sincerely Congratulations!!

The paper "Multi-Objective based Spatio-Temporal Feature Representation Learning Robust to Expression Intensity Variations for Facial Expression Recognition" has been accepted to IEEE Transactions on Affective Computing.

This paper was written by Dae Hoe Kim, Wisam J. Baddar and Yong Man Ro.

Seong Tae Kim and Hak Gu Kim Ph.D in IVY lab under the supervision of Prof. Yong Man Ro, received excellence award in the evaluation of research performance.

This award is presented to the graduate student with the excellent research accomplishments in the School of Electrical Engineering, KAIST.