Seoung-Jun Oh

Prof. Seoung-Jun Oh

Kwangwoon University
orcid

Research Focus & Metrics

Academic Career Length

19Years (2008-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 April 2021 in Remote Sensing
Verified
Journal Article
Open access
PDF file
WLWoo Ju Lee
Woo Ju Lee
DSDonggyu Sim
Donggyu Sim
SOSeoung-Jun Oh
Seoung-Jun Oh
In this paper, a convolutional neural network-based registration framework is proposed for remote sensing to improve the registration accuracy between two remote-sensed images acquired from different times and viewpoints. The proposed framework consists of four stages. In the first stage, key-points are extracted from two input images—a reference and a sensed image. Then, a patch is constructed at each key-point. The second stage consists of three processes for patch matching—candidate patch pair list generation, one-to-one matched label selection, and geometric distortion compensation. One-to-one matched patch pairs between two images are found, and the exact matching is found by compensating for geometric distortions in the matched patch pairs. A global geometric affine parameter set is computed using the random sample consensus algorithm (RANSAC) algorithm in the third stage. Finally, a registered image is generated after warping the input sensed image using the affine parameter set. The proposed high-accuracy registration framework is evaluated using the KOMPSAT-3 dataset by comparing the conventional frameworks based on machine learning and deep-learning-based frameworks. The proposed framework obtains the least root mean square error value of 34.922 based on all control points and achieves a 68.4% increase in the matching accuracy compared with the conventional registration framework.

2 August 2015 in Journal of Real-Time Image Processing
Verified
Journal Article
Closed access
PDF file
DLDongkyu Lee
Dongkyu Lee
DSDonggyu Sim
Donggyu Sim
KCKeeseong Cho
Keeseong Cho
SOSeoung-Jun Oh
Seoung-Jun Oh

9 March 2023 in IEEE Access
Verified
Journal Article
Open access
MLMinhun Lee
Minhun Lee
HCHansol Choi
Hansol Choi
JKJihoon Kim
Jihoon Kim
JDJihoon Do
Jihoon Do
HKHyoungjin Kwon
Hyoungjin Kwon
SOSeoung-Jun Oh
Seoung-Jun Oh
This paper presents a method to effectively compress the intermediate layer feature map of a convolutional neural network for the potential structures of Video Coding for Machines, which is an emerging technology for future machine consumption applications. Notably, most extant studies compress a single feature map and hence cannot entirely consider both global and local information within the feature map. This limits performance maintenance during machine consumption tasks that analyze objects with various sizes in images/videos. To address this problem, a multiscale feature map compression method is proposed that consists of two major processes: receptive block based principal component analysis (RPCA) and uniform integer quantization. The RPCA derives the complete basis kernels of a feature map by selecting a set of major basis kernels that can represent a sufficient percentage of global or local information according to the variable-size receptive blocks of each feature map. After transforming each feature map using the set of major basis kernels, a uniform integer quantizer converts the 32-bit floating-point values of the set of major basis kernels, corresponding RPCA coefficients, and a mean vector to five-bit integer representation values. Experiment results reveal that the proposed method reduces the amount of feature maps by 99.30% with a loss of 8.30% in the average precision ( AP ) on the OpenImageV6 dataset and 0.77% in APM and 0.47% in APL on the MS COCO 2017 validation set while outperforming previous PCA-based feature map compression methods even at higher compression rates.

30 November 2013 in Journal of Broadcast Engineering
Verified
Journal Article
Open access
YAYong-Jo Ahn
Yong-Jo Ahn
THTae-Jin Hwang
Tae-Jin Hwang
DLDongkyu Lee
Dongkyu Lee
SKSangmin Kim
Sangmin Kim
SOSeoung-Jun Oh
Seoung-Jun Oh
DSDonggyu Sim
Donggyu Sim

26 September 2023 in Electronics
Verified
Journal Article
Open access
PDF file
MLMinhun Lee
Minhun Lee
SPSeungjin Park
Seungjin Park
SOSeoung-Jun Oh
Seoung-Jun Oh
YKYounhee Kim
Younhee Kim
SJSe Yoon Jeong
Se Yoon Jeong
DSDonggyu Sim
Donggyu Sim
The burgeoning field of machine vision has led to the development by the Moving Picture Experts Group (MPEG) of a new type of compression technology called video coding for machines (VCM), to enhance machine recognition through video information compression. This research proposes a principal component analysis (PCA)-based compression methodology for multi-level feature maps extracted from the feature pyramid network (FPN) structure. Unlike current PCA-based studies that independently carry out PCA for each feature map, our approach employs a generalized basis matrix and mean vector derived from channel correlations by a generalized PCA process to eliminate the need for a PCA process. Further compression is achieved by amalgamating high-dimensional feature maps, capitalizing on the spatial redundancy within these multi-level feature maps. As a result, the proposed VCM encoder forgoes the PCA process, and the generalized data do not incur any compression loss. It only requires compressing the coefficients for each feature map using versatile video coding (VVC). Experimental results demonstrate superior performance by our method over all feature anchors for each machine vision task, as specified by the MPEG-VCM common test conditions, outperforming previous PCA-based feature map compression methods. Notably, it achieved an 89.3% BD-rate reduction for instance segmentation tasks.

15 July 2008 in IEEE Transactions on Consumer Electronics
Verified
Journal Article
Closed access
SYSang-Jun Yu
Sang-Jun Yu
JMJin-Su Myung
Jin-Su Myung
DSDonggyu Sim
Donggyu Sim
SOSeoung-Jun Oh
Seoung-Jun Oh
In this paper, we proposed an enhanced transform-domain intra prediction method for H.264/AVC. In the proposed scheme, H.264/AVC intra prediction can be described by generalized linear operation after defining both H-filter and V-filter. Since intra prediction operation can be decomposed into some sparse matrices, we can reduce computational complexity for intra prediction. Our scheme can reduce the number of multiplication and addition operations more than Chen's by 22.4% and 21.6%, respectively, in each 4times4 block. Furthermore, we show that matrix multiplication can be simply processed using a look-up table which consists of 9 types of 4times4 constant matrices corresponding to 9 prediction modes in H.264. Consequently, we can insist that the proposed scheme in this paper can be used for fast transform domain intra prediction in the area of transcoding MPEG-2 streams into H.264/AVC streams.