no code implementations • 29 Jan 2024 • Jaewoo Park, Jaeguk Kim, Nam Ik Cho
Accurately estimating the pose of an object is a crucial task in computer vision and robotics.
1 code implementation • ICCV 2023 • Haesoo Chung, Nam Ik Cho
In this paper, we propose an end-to-end HDR video composition framework, which aligns LDR frames in the feature space and then merges aligned features into an HDR frame, without relying on pixel-domain optical flow.
1 code implementation • 22 Aug 2023 • Haesoo Chung, Nam Ik Cho
For HDR imaging, some methods capture multiple low dynamic range (LDR) images with altering exposures to aggregate more information.
no code implementations • ICCV 2023 • Yeong Il Jang, Keuntek Lee, Gu Yong Park, Seyun Kim, Nam Ik Cho
There have been many image denoisers using deep neural networks, which outperform conventional model-based methods by large margins.
no code implementations • 21 Mar 2023 • Hochang Rhee, Seyun Kim, Nam Ik Cho
The decoder is constructed with corresponding video/image decoders and a new restoration network, which enhances the compressed video in two-step processes.
1 code implementation • CVPR 2023 • Seung Ho Park, Young Su Moon, Nam Ik Cho
Specifically, the framework comprises two models: a predictive model that infers an optimal objective map for a given low-resolution (LR) input and a generative model that applies a target objective map to produce the corresponding SR output.
Ranked #1 on Image Super-Resolution on DIV2K val - 4x upscaling
1 code implementation • 3 Jul 2022 • Jae Woong Soh, Nam Ik Cho
This paper presents a new variational inference framework for image restoration and a convolutional neural network (CNN) structure that can solve the restoration problems described by the proposed framework.
no code implementations • 3 Jul 2022 • Jae Woong Soh, Nam Ik Cho
Eventually, we propose a guideline for the patch extraction from given training images.
no code implementations • 26 May 2022 • Wonjun Kang, Geonsu Lee, Hyung Il Koo, Nam Ik Cho
The goal of face reenactment is to transfer a target expression and head pose to a source face while preserving the source identity.
1 code implementation • 13 Jan 2022 • Seung Ho Park, Young Su Moon, Nam Ik Cho
Instead of using multiple models, we present a more efficient method to train a single adjustable SR model on various combinations of losses by taking advantage of multi-task learning.
Ranked #1 on Image Super-Resolution on General100 - 8x upscaling
1 code implementation • 16 Dec 2021 • Young Kyun Jang, Geonmo Gu, Byungsoo Ko, Isaac Kang, Nam Ik Cho
To mitigate this issue, data augmentation can be applied during training.
1 code implementation • 16 Dec 2021 • Jaewoo Park, Nam Ik Cho
Our pose estimation method, dynamic projective spatial transformer network (DProST), localizes the region of interest grid on the rays in camera space and transforms the grid to object space by estimated pose.
no code implementations • CVPR 2022 • Hochang Rhee, Yeong Il Jang, Seyun Kim, Nam Ik Cho
Recent learning-based lossless image compression methods encode an image in the unit of subimages and achieve comparable performances to conventional non-learning algorithms.
1 code implementation • 8 Dec 2021 • Karam Park, Jae Woong Soh, Nam Ik Cho
We also propose a residual self-attention (RSA) module to further boost the performance, which produces 3-dimensional attention maps without additional parameters by cooperating with residual structures.
1 code implementation • ICCV 2021 • Young Kyun Jang, Nam Ik Cho
Supervised deep learning-based hash and vector quantization are enabling fast and large-scale image retrieval systems.
no code implementations • 11 Jul 2021 • Young Kyun Jang, Nam Ik Cho
Face image retrieval, which searches for images of the same identity from the query input face image, is drawing more attention as the size of the image database increases rapidly.
no code implementations • 19 Apr 2021 • Joon Young Ahn, Nam Ik Cho
The recent progress of deep convolutional neural networks has enabled great success in single image super-resolution (SISR) and many other vision tasks.
1 code implementation • 2 Apr 2021 • Jae Woong Soh, Nam Ik Cho
These methods separate the original problem into easier sub-problems and thus have shown improved performance than the naively trained CNN.
1 code implementation • 18 Jan 2021 • Jae Woong Soh, Nam Ik Cho
Traditionally, many researchers have investigated image priors for the denoising, within the Bayesian perspective based on image properties and statistics.
2 code implementations • CVPR 2020 • Jae Woong Soh, Sunwoo Cho, Nam Ik Cho
Despite their remarkable performance based on the external dataset, they cannot exploit internal information within a specific image.
1 code implementation • CVPR 2020 • Yoonsik Kim, Jae Woong Soh, Gu Yong Park, Nam Ik Cho
Real-noise denoising is a challenging task because the statistics of real-noise do not follow the normal distribution, and they are also spatially and temporally changing.
Ranked #12 on Image Denoising on DND (using extra training data)
2 code implementations • CVPR 2020 • Young Kyun Jang, Nam Ik Cho
Image retrieval methods that employ hashing or vector quantization have achieved great success by taking advantage of deep learning.
no code implementations • 3 Dec 2019 • Sungkwon Choo, Wonkyo Seo, Nam Ik Cho
The two-stream fusion network again consists of motion and appearance stream networks, which extract long-term temporal and spatial information, respectively.
1 code implementation • CVPR 2019 • Jae Woong Soh, Gu Yong Park, Junho Jo, Nam Ik Cho
Recently, many convolutional neural networks for single image super-resolution (SISR) have been proposed, which focus on reconstructing the high-resolution images in terms of objective distortion measures.
no code implementations • 8 Jul 2019 • Sepidehsadat Hosseini, Mohammad Amin Shabani, Nam Ik Cho
We propose a new semi-supervised learning method on face-related tasks based on Multi-Task Learning (MTL) and data distillation.
1 code implementation • 12 Jun 2019 • Junho Jo, Hyung Il Koo, Jae Woong Soh, Nam Ik Cho
For training our network, we develop a cross-entropy based loss function that addresses the imbalance problems.
1 code implementation • 2 May 2019 • Jae Woong Soh, Jae Sung Park, Nam Ik Cho
This paper presents a new framework for jointly enhancing the resolution and the dynamic range of an image, i. e., simultaneous super-resolution (SR) and high dynamic range imaging (HDRI), based on a convolutional neural network (CNN).
no code implementations • 2 Mar 2019 • Uiwon Hwang, Jaewoo Park, Hyemi Jang, Sungroh Yoon, Nam Ik Cho
Deep neural networks are widely used and exhibit excellent performance in many areas.
Ranked #2 on Adversarial Defense against FGSM Attack on MNIST
no code implementations • 24 Jan 2018 • Sepidehsadat Hosseini, Seok Hee Lee, Nam Ik Cho
In this paper, we show that finding an appropriate feature for the given problem may be still important as they can en- hance the performance of CNN-based algorithms.
1 code implementation • 2 Aug 2017 • Jae Sung Park, Nam Ik Cho
This paper presents an algorithm that enhances undesirably illuminated images by generating and fusing multi-level illuminations from a single image. The input image is first decomposed into illumination and reflectance components by using an edge-preserving smoothing filter.
no code implementations • 29 Jun 2017 • Dong-ju Jeong, Insung Hwang, Nam Ik Cho
We utilize deep saliency networks to transfer co-saliency prior knowledge and better capture high-level semantic information, and the resulting initial co-saliency maps are enhanced by seed propagation steps over an integrated graph.
no code implementations • 12 May 2017 • Byeongyong Ahn, Nam Ik Cho
There have been many discriminative learning methods using convolutional neural networks (CNN) for several image restoration problems, which learn the mapping function from a degraded input to the clean output.
no code implementations • 3 Apr 2017 • Byeongyong Ahn, Nam Ik Cho
There are two main streams in up-to-date image denoising algorithms: non-local self similarity (NSS) prior based methods and convolutional neural network (CNN) based methods.
no code implementations • 22 Jan 2017 • Yoonsik Kim, Insung Hwang, Nam Ik Cho
From these observations, for enjoying the performance of inception-like structure on the image based problems we propose a new convolutional network-in-network structure.