Main Page

From CV3VPL
Jump to: navigation, search

Welcome to Computer Vision & 3-D Video Processing Lab.

CV3VP Lab., advised by Prof. Kang-Sun Choi, was founded in 2011.

The major research field of CV3VP lab. covers cutting-edge computer vision algorithms and 3D video processing technologies including

3D Processing
/ Computer Vision

Image Processing
/ Understanding

Video Processing

  • Polygon mesh processing
  • Motion estimation
  • Disparity estimation (depth estimation)
  • Feature descriptor
  • Image retargeting
  • Image denoising
  • Wavelets
  • Color reproduction for CVD
  • Motion estimation
  • Video compression (H.264)

Join our Lab. page in Facebook !

News

"SLIC acceleration by early candidate cluster exclusion", Journal of Real-Time Imaging, May 2016

The manuscripts entitled "Acceleration of Simple Linear Iterative Clustering using Early Candidate Cluster Exclusion" was published in Journal of Real-Time Imaging. (SCIE, IF:2.020, 2015) (Article).

By applying the Cauchy-Schwarz inequality, an approximate distance metric for superpixel segmentation is derived, which allows unnecessary computation to be got rid of from the cluster inspection procedure. The proposed superpixel segmentation algorithm improves SLIC's computational efficiency by 219%, on average, without any degradation in segmentation accuracy.

"Fast SLIC using the subsampling technique", Computer Vision Image Understanding, May 2016

The manuscripts entitled "Subsampling-based Acceleration of Simple Linear Iterative Clustering for Superpixel Segmentation" was published in Computer Vision and Image Understanding. (SCI, IF:1.540, 2015) (Article).

Exploiting high interpixel redundancy of images, the proposed subsampling approach to superpixel segmentation boosts performance of SLIC up to five times.

Electronics Letters, Nov. 2013

The manuscript entitled "Angular-Partitioned Spin Image Descriptor for Robust 3D Facial Landmark Detection" was published in Electronics Letters. (SCI, IF: 1.038, 2012) (Article).

The proposed descriptor improves uniqueness of features in representing surface shapes of 3D point clouds.

Digital Signal Processing, July 2013

The manuscript entitled "Bit Plane Modification for Improving MSE-Near Optimal DPCM-Based Block Truncation Coding" was published in Digital Signal Processing. (SCI, IF: 1.918, 2012) (Article).

Workshop on Image Understanding, Kyung Sung University, Jan. 28, 2013

Prof. Choi gave a talk on Content-Aware Image Retargeting at the Workshop on Image Understanding (Kyung Sung University, Pusan) on January 28th, 2013.

The contents of the talk are the following topics:

  • Perceptual importance evaluation
  • Retargeting operators
  • Different approaches
  • Quality assessment
  • Applications
IEIE Achievement Award, Nov. 29, 2014

Prof. Choi won the achievement award from The Institute of Electronics and Information Engineers on November 29th, 2014.

He has been serving as the Journal coordinator of International Journal of Smart Processing and Computing in IEIE.

Content-Aware 3D Image Retargeting, Optical Engineering, 2012

A content-aware image retargeting algorithm was published in the April issue of Optical Engineering in 2012. (Article)

The proposed algorithm preserves surfaces of 3D objects after retargeting 3D images by performing retargeting the 3D data not only in the spatial coordinates but also in the depth information.

Special Merit Award, IEEE CE Society, 2012

Prof. Choi won the special merit award at IEEE International Conference on Consumer Electronics on January 15th, 2012.

Motion estimation utilizing the disparity characteristics of stereoscopic videos was evaluated as an outstanding achievement.

The related patent 12-0054075 was also applied on May 22, 2012.



"KOREATECH"
"School of EECE"




Whos here now:   Members 0   Guests 1   Bots & Crawlers 1
 
Personal tools
Namespaces

Variants
Actions
People
Project
Publication
Lecture
Toolbox