Statistical Learning for Signal Processing Lab. (SLSP) 

(formally Multimedia Processing Lab.) was established under the guidance of Professor Chang D. Yoo in 1999, the year he arrived at KAIST. Using various machine learning theories and novel signal processing techniques, signals such as image, text, speech, audio, video, EEG and financial data are processed for longstanding and emerging applications.  


For the Direction to Our Lab, Please Click Here.


 [Lab. Introduction]

SLSP Lab. News

more >
Team KAIST-SLSP participated in Imagenet Large Scale Visual Recognition Challenge 2016 (ILSVRC 2016) and ranked 7th place in object detection part (DET), and 5th pla...
IEIE Workshop 'Theory and Exercise in Deep Learning' successfully took place in Multi-purpose Hall, IT Convergence Building(N1),KAIST, Daejeon, Korea through Apr. 29– 30...
'Machine Learning Tutorial 2016’ successfully took place in Multi-purpose Hall, IT Convergence Building(N1), KAIST, Daejeon, Korea through Mar. 24 – 27,2016. This tutorial...
SLSP has two graduates this February, who have spent memorable times and developed their research capacity for two years. Having made several achievements, MS cand...

Recently Accepted Papers

paper

ICASSP2017.png

Hyunsin Park, Chang D. Yoo, "Melody extraction and detection through LSTM-RNN with harmonic sum loss", in Proceedings of International Conference on Acoustics, Speech and Signal Processing, New Orleans, USA, 2017

ICIP2016.png

Mingoo Song, Chang D. Yoo, "Multimodal Representation: Kneser-Ney Smoothing/Skip-gram based Neural Language Model", in Proceedings of International Conference on Image Processing, Phoenix, USA, 2016.

Interspeech.png

Haeyong Kang, Chang D. Yoo, and Yongcheon Na, "Maximum Margin Learning of t-SPNs for Cell Classification with Filtered Input", IEEE Journal of Selected Topics in Signal Processing, vol.10, no.1, pp.130-139, February 2016.

Interspeech.png

Sanghyuk Park, Hyunsin Park, and Chang D. Yoo, "Complex Video Scene Analysis using Kernelized Collaborative Behavior Pattern Learning based on Hierarchical Representative Object Behaviors", IEEE Transactions on Circuits and Systems for Video Technology, accepted for publication.