Search Results for author: Soo-Chang Pei

Found 7 papers, 1 papers with code

Domain Adaptation for Underwater Image Enhancement via Content and Style Separation

1 code implementation17 Feb 2022 Yu-Wei Chen, Soo-Chang Pei

Underwater image suffer from color cast, low contrast and hazy effect due to light absorption, refraction and scattering, which degraded the high-level application, e. g, object detection and object tracking.

Domain Adaptation Image Enhancement +4

Explorable Tone Mapping Operators

no code implementations20 Oct 2020 Chien-Chuan Su, Ren Wang, Hung-Jin Lin, Yu-Lun Liu, Chia-Ping Chen, Yu-Lin Chang, Soo-Chang Pei

It aims to preserve visual information of HDR images in a medium with a limited dynamic range.

Tone Mapping

Difference-Seeking Generative Adversarial Network--Unseen Sample Generation

no code implementations ICLR 2020 Yi Lin Sung, Sung-Hsien Hsieh, Soo-Chang Pei, Chun-Shien Lu

Unseen data, which are not samples from the distribution of training data and are difficult to collect, have exhibited importance in numerous applications, ({\em e. g.,} novelty detection, semi-supervised learning, and adversarial training).

Difference-Seeking Generative Adversarial Network

no code implementations ICLR 2019 Yi-Lin Sung, Sung-Hsien Hsieh, Soo-Chang Pei, Chun-Shien Lu

DSGAN considers the scenario that the training samples of target distribution, $p_{t}$, are difficult to collect.

Discrete Gyrator Transforms: Computational Algorithms and Applications

no code implementations3 Jun 2017 Soo-Chang Pei, Shih-Gu Huang, Jian-Jiun Ding

Besides, we propose a kind of DGT based on the eigenfunctions of the gyrator transform.

Two-dimensional nonseparable discrete linear canonical transform based on CM-CC-CM-CC decomposition

no code implementations26 May 2017 Soo-Chang Pei, Shih-Gu Huang

Simulation results show that the proposed methods have higher accuracy, lower computational complexity and smaller error in the additivity property compared with the previous works.

Fast Binary Embedding via Circulant Downsampled Matrix -- A Data-Independent Approach

no code implementations24 Jan 2016 Sung-Hsien Hsieh, Chun-Shien Lu, Soo-Chang Pei

Binary embedding of high-dimensional data aims to produce low-dimensional binary codes while preserving discriminative power.

Cannot find the paper you are looking for? You can Submit a new open access paper.