no code implementations • Findings (EMNLP) 2021 • Fuxiang Chen, Mijung Kim, Jaegul Choo
To tackle this problem, previous work on code summarization, the task of automatically generating code description given a piece of code reported that an auxiliary learning model trained to produce API (Application Programming Interface) embeddings showed promising results when applied to a downstream, code summarization model.
1 code implementation • 22 Dec 2024 • Jeongho Kim, Hoiyeong Jin, Sunghyun Park, Jaegul Choo
In the text-editable virtual try-on, three key aspects exist: (i) designing rich text descriptions for paired person-clothing data to train the model, (ii) addressing the conflicts where textual information of the existing person's clothing interferes the generation of the new clothing, and (iii) adaptively adjust the inpainting mask aligned with the text descriptions, ensuring proper editing areas while preserving the original person's appearance irrelevant to the new clothing.
no code implementations • 18 Dec 2024 • Sanghyeon Lee, Jooyeol Yun, Jaegul Choo
Point-based interactive colorization techniques allow users to effortlessly colorize grayscale images using user-provided color hints.
1 code implementation • 6 Dec 2024 • Hyesu Lim, Jinho Choi, Jaegul Choo, Steffen Schneider
Adapting foundation models for specific purposes has become a standard approach to build machine learning systems for downstream applications.
no code implementations • 27 Nov 2024 • Junha Hyung, Kinam Kim, Susung Hong, Min-Jung Kim, Jaegul Choo
Our contributions include: (1) introducing STG as an efficient, high-performing guidance technique for video diffusion models, (2) eliminating the need for auxiliary models by simulating a weak model through layer skipping, and (3) ensuring quality-enhanced guidance without compromising sample diversity or dynamics unlike CFG.
1 code implementation • 21 Nov 2024 • Daehoon Gwak, Junwoo Park, Minho Park, ChaeHun Park, Hyunchan Lee, Edward Choi, Jaegul Choo
Predicting future international events from textual information, such as news articles, has tremendous potential for applications in global policy, strategic decision-making, and geopolitics.
no code implementations • 24 Oct 2024 • ChaeHun Park, Hojun Cho, Jaegul Choo
This paper explores integrating Automatic Speech Recognition (ASR) into natural language query systems to improve weather forecasting efficiency for Korean meteorologists.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+4
no code implementations • 17 Oct 2024 • Minseok Choi, ChaeHun Park, Dohyun Lee, Jaegul Choo
Large language models (LLMs) serve as giant information stores, often including personal or copyrighted data, and retraining them from scratch is not a viable option.
no code implementations • 17 Oct 2024 • Jooyeol Yun, Davide Abati, Mohamed Omran, Jaegul Choo, Amirhossein Habibian, Auke Wiggers
In this paper, we focus on the former, creating a location model dedicated to identifying realistic object locations.
no code implementations • 15 Oct 2024 • Jaeseong Lee, Taewoong Kang, Marcel C. Bühler, Min-Jung Kim, Sungwon Hwang, Junha Hyung, Hyojin Jang, Jaegul Choo
Recent advancements in head avatar rendering using Gaussian primitives have achieved significantly high-fidelity results.
3 code implementations • 13 Oct 2024 • Hojoon Lee, Dongyoon Hwang, Donghu Kim, Hyunseung Kim, Jun Jet Tai, Kaushik Subramanian, Peter R. Wurman, Jaegul Choo, Peter Stone, Takuma Seno
Recent advances in CV and NLP have been largely driven by scaling up the number of network parameters, despite traditional theories suggesting that larger networks are prone to overfitting.
no code implementations • 13 Oct 2024 • Soyoung Yang, Hojun Cho, Jiyoung Lee, Sohee Yoon, Edward Choi, Jaegul Choo, Won Ik Cho
Aspect-based sentiment analysis (ABSA) is the challenging task of extracting sentiment along with its corresponding aspects and opinions from human language.
Aspect-Based Sentiment Analysis
Aspect-Based Sentiment Analysis (ABSA)
+2
no code implementations • 5 Sep 2024 • Jinhee Kim, Taesung Kim, Jaegul Choo
Recent advances in interactive keypoint estimation methods have enhanced accuracy while minimizing user intervention.
no code implementations • 4 Sep 2024 • Kyungmin Jo, Jaegul Choo
However, depth-based pose control confronts issues of shape dependency, as depth maps influence not only the pose but also the shape of the generated images.
1 code implementation • 29 Aug 2024 • Chaeyeon Chung, Sunghyun Park, Jeongho Kim, Jaegul Choo
Hairstyle transfer is a challenging task in the image editing field that modifies the hairstyle of a given face image while preserving its other appearance and background features.
no code implementations • 15 Aug 2024 • Wonwoo Cho, Kangyeol Kim, Saemee Choi, Jaegul Choo
Despite the growing prevalence of black-box pre-trained models (PTMs) such as prediction API services, there remains a significant challenge in directly applying general models to real-world scenarios due to the data distribution gap.
1 code implementation • 22 Jul 2024 • Junha Song, Tae Soo Kim, Junha Kim, Gunhee Nam, Thijs Kooi, Jaegul Choo
This approach helps existing SemiSDA methods to adapt the model with a balanced supervised signal by utilizing latent defending samples throughout the adaptation process.
1 code implementation • 15 Jul 2024 • Chung Park, Taesan Kim, Hyungjun Yoon, Junui Hong, Yelim Yu, Mincheol Cho, Minsung Choi, Jaegul Choo
To this end, our model compares the performance of a model trained on multiple domains (CDSR) with a model trained solely on the specific domain (SDSR) to evaluate the negative transfer of each domain using our asymmetric cooperative network.
no code implementations • 13 Jul 2024 • Kangyeol Kim, Wooseok Seo, Sehyun Nam, Bodam Kim, Suhyeon Jeong, Wonwoo Cho, Jaegul Choo, Youngjae Yu
Personalized text-to-image (P-T2I) generation aims to create new, text-guided images featuring the personalized subject with a few reference images.
no code implementations • 12 Jul 2024 • Jeongho Kim, Min-Jung Kim, Junsoo Lee, Jaegul Choo
Extensive experiments demonstrate that the proposed method can achieve promising results in video synthesis tasks encompassing various poses, like chibi.
1 code implementation • 9 Jul 2024 • Jooyeol Yun, Jaegul Choo
To overcome this long-standing scalability challenge, we present a unique approach that leverages readily available databases for general image aesthetic assessment and image quality assessment.
1 code implementation • 3 Jul 2024 • Sungwon Hwang, Min-Jung Kim, Taewoong Kang, Jayeon Kang, Jaegul Choo
To the best of our knowledge, we are the first to address the EVS problem in urban scene reconstruction.
no code implementations • 24 Jun 2024 • Yujin Baek, ChaeHun Park, Jaeseok Kim, Yu-Jung Heo, Du-Seong Chang, Jaegul Choo
We demonstrate the effectiveness of this framework through the creation of K-Viscuit, a dataset focused on Korean culture.
no code implementations • 20 Jun 2024 • Dohyun Lee, Daniel Rim, Minseok Choi, Jaegul Choo
In this work, we propose Privacy Protection via Optimal Parameters (POP), a novel unlearning method that effectively forgets the target token sequences from the pretrained LM by applying optimal gradient updates to the parameters.
no code implementations • 18 Jun 2024 • Minseok Choi, Daniel Rim, Dohyun Lee, Jaegul Choo
Instruction-following large language models (LLMs), such as ChatGPT, have become widely popular among everyday users.
no code implementations • 18 Jun 2024 • Hawon Jeong, ChaeHun Park, Jimin Hong, Hojoon Lee, Jaegul Choo
As large language models (LLMs) are increasingly used as evaluators for natural language generation tasks, ensuring unbiased assessments is essential.
no code implementations • 18 Jun 2024 • Seungbin Yang, ChaeHun Park, Taehee Kim, Jaegul Choo
However, these tool-augmented LLMs often encounter incomplete scenarios when users provide partial information or the necessary tools are unavailable.
1 code implementation • 18 Jun 2024 • Minseok Choi, Kyunghyun Min, Jaegul Choo
Pretrained language models memorize vast amounts of information, including private and copyrighted data, raising significant safety concerns.
no code implementations • 17 Jun 2024 • Junha Hyung, Susung Hong, Sungwon Hwang, Jaeseong Lee, Jaegul Choo, Jin-Hwa Kim
3D reconstruction from multi-view images is one of the fundamental challenges in computer vision and graphics.
1 code implementation • 10 Jun 2024 • Dongyoon Hwang, Byungkun Lee, Hojoon Lee, Hyunseung Kim, Jaegul Choo
To this end, we introduce Convolution Injector (CoIn), an add-on module that injects convolutions which are rich in locality and equivariance biases into a pretrained ViT for effective adaptation in visuo-motor control.
1 code implementation • 10 Jun 2024 • Donghu Kim, Hojoon Lee, Kyungmin Lee, Dongyoon Hwang, Jaegul Choo
In contrast, objectives focused on learning task-specific knowledge (e. g., identifying agents and fitting reward functions) improve performance in environments similar to the pre-training dataset but not in varied ones.
1 code implementation • 8 Jun 2024 • Minho Park, Sunghyun Park, Jooyeol Yun, Jaegul Choo
However, despite the high fidelity of generated images, we observed a significant performance degradation when fine-tuning the model using the generated datasets due to the domain gap between real and generated images.
no code implementations • 4 Jun 2024 • ChaeHun Park, Koanho Lee, Hyesu Lim, Jaeseok Kim, Junmo Park, Yu-Jung Heo, Du-Seong Chang, Jaegul Choo
Building a reliable visual question answering~(VQA) system across different languages is a challenging problem, primarily due to the lack of abundant samples for training.
no code implementations • 1 Jun 2024 • Hyunseung Kim, Byungkun Lee, Hojoon Lee, Dongyoon Hwang, Donghu Kim, Jaegul Choo
First, in an instruction learning stage, DoDont leverages action-free instruction videos to train an instruction network to distinguish desirable transitions from undesirable ones.
1 code implementation • 1 Jun 2024 • Hojoon Lee, Hyeonseo Cho, Hyunseung Kim, Donghu Kim, Dugki Min, Jaegul Choo, Clare Lyle
By periodically reinitializing the Hare network to the Tortoise's weights, our method preserves plasticity while retaining general knowledge.
no code implementations • CVPR 2024 • Jeonghoon Park, Chaeyeon Chung, Juyoung Lee, Jaegul Choo
In the image classification task, deep neural networks frequently rely on bias attributes that are spuriously correlated with a target class in the presence of dataset bias, resulting in degraded performance when applied to data without bias attributes.
1 code implementation • 15 Apr 2024 • Jinhee Kim, Taesung Kim, Jaegul Choo
In this work, we explore the effectiveness of LLMs for generating realistic synthetic tabular data, identifying key prompt design elements to optimize performance.
1 code implementation • 1 Apr 2024 • ChaeHun Park, Minseok Choi, Dohyun Lee, Jaegul Choo
Recent studies proposed evaluation metrics that assess generated responses by considering their relevance to previous dialogue histories.
no code implementations • 12 Feb 2024 • Jaeseong Lee, Junha Hyung, SOHYUN JEONG, Jaegul Choo
Most previous face swapping approaches have relied on the seesaw game training scheme, also known as the target-oriented approach.
1 code implementation • 3 Feb 2024 • Junwoo Park, Daehoon Gwak, Jaegul Choo, Edward Choi
To this end, our contrastive loss incorporates global autocorrelation held in the whole time series, which facilitates the construction of positive and negative pairs in a self-supervised manner.
Ranked #1 on
Time Series Forecasting
on ETTh1 (720) Univariate
no code implementations • 24 Dec 2023 • Dongmin Choi, Wonwoo Cho, Kangyeol Kim, Jaegul Choo
Accurately annotating multiple 3D objects in LiDAR scenes is laborious and challenging.
1 code implementation • 19 Dec 2023 • Dongmin Kim, Sunghyun Park, Jaegul Choo
Time-series anomaly detection deals with the problem of detecting anomalous timesteps by learning normality from the sequence of observations.
no code implementations • 6 Dec 2023 • Chung Park, Taekyoon Choi, Taesan Kim, Mincheol Cho, Junui Hong, Minsung Choi, Jaegul Choo
Previous studies using large-scale trajectory datasets in a single server have achieved remarkable performance in UNLP task.
1 code implementation • CVPR 2024 • Jeongho Kim, Gyojung Gu, Minho Park, Sunghyun Park, Jaegul Choo
Given a clothing image and a person image, an image-based virtual try-on aims to generate a customized image that appears natural and accurately reflects the characteristics of the clothing image.
1 code implementation • 22 Nov 2023 • Chung Park, Taesan Kim, Taekyoon Choi, Junui Hong, Yelim Yu, Mincheol Cho, Kyunam Lee, Sungil Ryu, Hyungjun Yoon, Minsung Choi, Jaegul Choo
This paper investigates Cross-Domain Sequential Recommendation (CDSR), a promising method that uses information from multiple domains (more than three) to generate accurate and diverse recommendations, and takes into account the sequential nature of user interactions.
1 code implementation • 3 Nov 2023 • Changdae Oh, Hyesu Lim, Mijoo Kim, Dongyoon Han, Sangdoo Yun, Jaegul Choo, Alexander Hauptmann, Zhi-Qi Cheng, Kyungwoo Song
Improving out-of-distribution (OOD) generalization during in-distribution (ID) adaptation is a primary goal of robust fine-tuning of zero-shot models beyond naive fine-tuning.
1 code implementation • 16 Oct 2023 • Taewoong Kang, Jeongsik Oh, Jaeseong Lee, Sunghyun Park, Jaegul Choo
Specifically, to maintain the geometric consistency of expressions between the input and output of the expression domain translation network, we employ a 3D geometric-aware loss function that reduces the distances between the vertices in the 3D mesh of the human and anime.
1 code implementation • 12 Oct 2023 • Minseok Choi, Chaeheon Gwak, SeHo Kim, Si Hyeong Kim, Jaegul Choo
Keyphrase generation (KG) aims to generate a set of summarizing words or phrases given a source document, while keyphrase extraction (KE) aims to identify them from the text.
no code implementations • 2 Oct 2023 • Chung Park, Junui Hong, Cheonbok Park, Taesan Kim, Minsung Choi, Jaegul Choo
Understanding the movement patterns of objects (e. g., humans and vehicles) in a city is essential for many applications, including city planning and management.
no code implementations • 2 Oct 2023 • Chung Park, Taesan Kim, Junui Hong, Minsung Choi, Jaegul Choo
To tackle this problem, we propose a Geo-Tokenizer, designed to efficiently reduce the number of locations to be trained by representing a location as a combination of several grids at different scales.
1 code implementation • 25 Sep 2023 • Minseok Choi, Hyesu Lim, Jaegul Choo
Document-level relation extraction (DocRE) aims to extract relations of all entity pairs in a document.
no code implementations • 22 Sep 2023 • Jimin Hong, ChaeHun Park, Jaegul Choo
We then enhance the diversity of the second model by focusing on patterns that the first model fails to learn.
no code implementations • 19 Sep 2023 • Kyungmin Jo, Wonjoon Jin, Jaegul Choo, Hyunjoon Lee, Sunghyun Cho
In this paper, we propose SideGAN, a novel 3D GAN training method to generate photo-realistic images irrespective of the camera pose, especially for faces of side-view angles.
no code implementations • 13 Sep 2023 • Junha Hyung, Jaeyo Shin, Jaegul Choo
The main challenge with this task is the absence of ground truth for the composed concepts, leading to a reduction in the quality of the final output and an identity shift of the source subject.
no code implementations • 7 Sep 2023 • Sungwon Hwang, Junha Hyung, Jaegul Choo
Our main strategy is to construct the 3D avatar in Neural Radiance Fields (NeRF) optimized with a set of controlled viewpoint-aware images that we generate from ControlNet, whose condition input is the depth map extracted from the input video.
no code implementations • 22 Aug 2023 • Hojoon Lee, Dongyoon Hwang, Kyushik Min, Jaegul Choo
In this work, we revisited experiments on IRS with review datasets and compared RL-based models with a simple reward model that greedily recommends the item with the highest one-step reward.
1 code implementation • 21 Aug 2023 • Hojoon Lee, Hawon Jeong, Byungkun Lee, Kyungyup Lee, Jaegul Choo
In this paper, we introduce ST-RAP, a novel Spatio-Temporal framework for Real estate APpraisal.
no code implementations • ICCV 2023 • Sunghyun Park, Seunghan Yang, Jaegul Choo, Sungrack Yun
Test-time adaptation (TTA) aims to adapt a pre-trained model to the target domain in a batch-by-batch manner during inference.
1 code implementation • ICCV 2023 • Minho Park, Jooyeol Yun, Seunghwan Choi, Jaegul Choo
Our experiments reveal that we can guide text-to-image generation models to be aware of the semantics of different image regions, by training the model to generate semantic labels for each pixel.
no code implementations • ICCV 2023 • Chaeyeon Chung, Yeojeong Park, Seunghwan Choi, Munkhsoyol Ganbat, Jaegul Choo
Shourcut-V2V avoids full inference for every neighboring video frame by approximating the intermediate features of a current frame from those of the previous frame.
1 code implementation • ICCV 2023 • Jungsoo Lee, Debasmit Das, Jaegul Choo, Sungha Choi
To be more specific, entropy minimization attempts to raise the confidence values of an individual sample's prediction, but individual confidence values may rise or fall due to the influence of signals from numerous other predictions (i. e., wisdom of crowds).
no code implementations • ICCV 2023 • Sungwon Hwang, Junha Hyung, Daejin Kim, Min-Jung Kim, Jaegul Choo
To do so, we first train a scene manipulator, a latent code-conditional deformable NeRF, over a dynamic scene to control a face deformation using the latent code.
no code implementations • 18 Jul 2023 • Gyumin Shim, Jaeseong Lee, Junha Hyung, Jaegul Choo
In this paper, we propose PixelHuman, a novel human rendering model that generates animatable human scenes from a few images of a person with unseen identity, views, and poses.
1 code implementation • 10 Jul 2023 • Soyoung Yang, Minseok Choi, Youngwoo Cho, Jaegul Choo
To demonstrate the usefulness of our dataset, we propose a bilingual RE model that leverages both Korean and Hanja contexts to predict relations between entities.
no code implementations • CVPR 2023 • Junha Hyung, Sungwon Hwang, Daejin Kim, Hyunji Lee, Jaegul Choo
Specifically, we present three add-on modules of LENeRF, the Latent Residual Mapper, the Attention Field Network, and the Deformation Network, which are jointly used for local manipulations of 3D features by estimating a 3D attention field.
1 code implementation • 21 Jun 2023 • Yujin Baek, Koanho Lee, Dayeon Ki, Hyoung-Gyu Lee, Cheonbok Park, Jaegul Choo
The effects of PLUMCOT are shown to be remarkable in "unseen" constraints.
1 code implementation • 9 Jun 2023 • Hojoon Lee, Koanho Lee, Dongyoon Hwang, Hyunho Lee, Byungkun Lee, Jaegul Choo
To address this issue, we propose a novel URL framework that causally predicts future states while increasing the dimension of the latent manifold by decorrelating the features in the latent space.
1 code implementation • 8 May 2023 • ChaeHun Park, Seungil Chad Lee, Daniel Rim, Jaegul Choo
Despite the recent advances in open-domain dialogue systems, building a reliable evaluation metric is still a challenging problem.
no code implementations • 31 Mar 2023 • Sunghyun Park, Sunghyo Chung, Jungsoo Lee, Jaegul Choo
However, STR models show a large performance degradation on languages with a numerous number of characters (e. g., Chinese and Korean), especially on characters that rarely appear due to the long-tailed distribution of characters in such languages.
1 code implementation • 31 Mar 2023 • Kangyeol Kim, Sunghyun Park, Junsoo Lee, Jaegul Choo
Recent remarkable improvements in large-scale text-to-image generative models have shown promising results in generating high-fidelity images.
no code implementations • 28 Mar 2023 • Jaeseong Lee, Taewoo Kim, Sunghyun Park, Younggun Lee, Jaegul Choo
However, we observed that previous approaches still suffer from source attribute leakage, where the source image's attributes interfere with the target image's.
1 code implementation • 9 Mar 2023 • Daeun Kyung, Kyungmin Jo, Jaegul Choo, Joonseok Lee, Edward Choi
X-ray computed tomography (CT) is one of the most common imaging techniques used to diagnose various diseases in the medical field.
1 code implementation • 27 Feb 2023 • Junwoo Park, Jungsoo Lee, Youngin Cho, Woncheol Shin, Dongmin Kim, Jaegul Choo, Edward Choi
Based on our findings, we propose a reweighting framework that down-weights the losses incurred by abrupt changes and up-weights those by normal states.
no code implementations • 10 Feb 2023 • Hyesu Lim, Byeonggeun Kim, Jaegul Choo, Sungha Choi
In this paper, we identify that CBN and TBN are in a trade-off relationship and present a new test-time normalization (TTN) method that interpolates the statistics by adjusting the importance between CBN and TBN according to the domain-shift sensitivity of each BN layer.
no code implementations • ICCV 2023 • Kyungmin Jo, Wonjoon Jin, Jaegul Choo, Hyunjoon Lee, Sunghyun Cho
In this paper, we propose SideGAN, a novel 3D GAN training method to generate photo-realistic images irrespective of the camera pose, especially for faces of side-view angles.
no code implementations • 9 Nov 2022 • Gyumin Shim, Minsoo Lee, Jaegul Choo
Single-image 3D human reconstruction aims to reconstruct the 3D textured surface of the human body given a single image.
no code implementations • 31 Oct 2022 • Nyoungwoo Lee, ChaeHun Park, Ho-Jin Choi, Jaegul Choo
To overcome these limitations, this paper proposes a simple but efficient method for generating adversarial negative responses leveraging a large-scale language model.
no code implementations • 25 Oct 2022 • Youngin Cho, Junsoo Lee, Soyoung Yang, Juntae Kim, Yeojeong Park, Haneol Lee, Mohammad Azam Khan, Daesik Kim, Jaegul Choo
Existing deep interactive colorization models have focused on ways to utilize various types of interactions, such as point-wise color hints, scribbles, or natural-language texts, as methods to reflect a user's intent at runtime.
no code implementations • 25 Oct 2022 • Youngin Cho, Daejin Kim, Dongmin Kim, Mohammad Azam Khan, Jaegul Choo
Time series forecasting has become a critical task due to its high practicality in real-world applications such as traffic, energy consumption, economics and finance, and disease analysis.
1 code implementation • 12 Oct 2022 • Junwoo Park, Youngwoo Cho, Gyuhyeon Sim, Hojoon Lee, Jaegul Choo
By exploiting the advantage of the game environment, we construct a gunshot dataset, namely BGG, for the firearm classification and gunshot localization tasks.
no code implementations • 21 Sep 2022 • Jihyeon Lee, Taehee Kim, Yunwon Tae, Cheonbok Park, Jaegul Choo
Incorporating personal preference is crucial in advanced machine translation tasks.
1 code implementation • 15 Sep 2022 • Jinhee Kim, Taesung Kim, Taewoo Kim, Jaegul Choo, Dong-Wook Kim, Byungduk Ahn, In-Seok Song, Yoon-Ji Kim
To fully automate this procedure, deep-learning-based methods have been widely proposed and have achieved high performance in detecting keypoints in medical images.
no code implementations • 14 Sep 2022 • Bum Chul Kwon, Jungsoo Lee, Chaeyeon Chung, Nyoungwoo Lee, Ho-Jin Choi, Jaegul Choo
We call the unwanted correlations "data biases," and the visual features causing data biases "bias factors."
no code implementations • 12 Sep 2022 • Daejin Kim, Youngin Cho, Dongmin Kim, Cheonbok Park, Jaegul Choo
Extensive experiments on METR-LA and PEMS-BAY demonstrate that our ResCAL can correctly capture the correlation of errors and correct the failures of various traffic forecasting models in event situations.
1 code implementation • COLING 2022 • Taehee Kim, ChaeHun Park, Jimin Hong, Radhika Dua, Edward Choi, Jaegul Choo
To analyze this, we first train a classifier that identifies machine-written sentences, and observe that the linguistic features of the sentences identified as written by a machine are significantly different from those of human-written sentences.
1 code implementation • 16 Aug 2022 • Taewoo Kim, Chaeyeon Chung, Yoonseo Kim, Sunghyun Park, Kangyeol Kim, Jaegul Choo
Editing hairstyle is unique and challenging due to the complexity and delicacy of hairstyle.
no code implementations • 8 Aug 2022 • Seungmin Jin, Hyunwook Lee, Cheonbok Park, Hyeshin Chu, Yunwon Tae, Jaegul Choo, Sungahn Ko
With deep learning (DL) outperforming conventional methods for different tasks, much effort has been devoted to utilizing DL in various domains.
no code implementations • 21 Jul 2022 • Wonwoo Cho, Jaegul Choo
To effectively solve the OSR problem, previous studies attempted to limit latent feature space and reject data located outside the limited space via offline analyses, e. g., distance-based feature analyses, or complicated network architectures.
1 code implementation • 14 Jul 2022 • Jooyeol Yun, Sanghyeon Lee, Minho Park, Jaegul Choo
It is essential for point-interactive colorization methods to appropriately propagate user-provided colors (i. e., user hints) in the entire image to obtain a reasonably colorized image with minimal user effort.
1 code implementation • 28 Jun 2022 • Sangyun Lee, Gyojung Gu, Sunghyun Park, Seunghwan Choi, Jaegul Choo
Image-based virtual try-on aims to synthesize an image of a person wearing a given clothing item.
Ranked #4 on
Virtual Try-on
on VITON-HD
no code implementations • 17 Jun 2022 • Chaeyeon Chung, Taewoo Kim, Hyelin Nam, Seunghwan Choi, Gyojung Gu, Sunghyun Park, Jaegul Choo
Hairstyle transfer is the task of modifying a source hairstyle to a target one.
no code implementations • 12 Jun 2022 • Youngin Cho, Daejin Kim, Mohammad Azam Khan, Jaegul Choo
Therefore, in this study we explore the practical setting called the single positive setting, where each data instance is annotated by only one positive label with no explicit negative labels.
no code implementations • 8 Jun 2022 • Jungsoo Lee, Juyoung Lee, Sanghun Jung, Jaegul Choo
Based on such issues, this paper 1) proposes an evaluation metric `Align-Conflict (AC) score' for the tuning criterion, 2) includes experimental settings with low bias severity and shows that they are yet to be explored, and 3) unifies the standardized experimental settings to promote fair comparisons between debiasing methods.
no code implementations • ICCV 2023 • Sanghun Jung, Jungsoo Lee, Nanhee Kim, Amirreza Shaban, Byron Boots, Jaegul Choo
That is, a model does not have a chance to learn test data in a class-discriminative manner, which was feasible in other adaptation tasks (\textit{e. g.,} unsupervised domain adaptation) via supervised losses on the source data.
1 code implementation • 29 May 2022 • Jungsoo Lee, Jeonghoon Park, Daeyoung Kim, Juyoung Lee, Edward Choi, Jaegul Choo
$f_B$ is trained to focus on bias-aligned samples (i. e., overfitted to the bias) while $f_D$ is mainly trained with bias-conflicting samples by concentrating on samples which $f_B$ fails to learn, leading $f_D$ to be less susceptible to the dataset bias.
1 code implementation • 27 Apr 2022 • Hojoon Lee, Dongyoon Hwang, Hyunseung Kim, Byungkun Lee, Jaegul Choo
To alleviate this problem, we propose DraftRec, a novel hierarchical model which recommends characters by considering each player's champion preferences and the interaction between the players.
no code implementations • 12 Mar 2022 • Minsoo Lee, Chaeyeon Chung, Hojun Cho, Minjung Kim, Sanghun Jung, Jaegul Choo, Minhyuk Sung
While NeRF-based 3D-aware image generation methods enable viewpoint control, limitations still remain to be adopted to various 3D applications.
no code implementations • 21 Dec 2021 • Kangyeol Kim, Sunghyun Park, Junsoo Lee, Joonseok Lee, Sookyung Kim, Jaegul Choo, Edward Choi
In order to perform unconditional video generation, we must learn the distribution of the real-world videos.
no code implementations • 7 Dec 2021 • Kyungmin Jo, Gyumin Shim, Sanghun Jung, Soyoung Yang, Jaegul Choo
While recent NeRF-based generative models achieve the generation of diverse 3D-aware images, these approaches have limitations when generating images that contain user-specified characteristics.
no code implementations • 16 Nov 2021 • Taewon Kang, Sunghyun Park, Seunghwan Choi, Jaegul Choo
Image-based virtual try-on provides the capacity to transfer a clothing item onto a photo of a given person, which is usually accomplished by warping the item to a given human pose and adjusting the warped item to the person.
1 code implementation • 15 Nov 2021 • Kangyeol Kim, Sunghyun Park, Jaeseong Lee, Sunghyo Chung, Junsoo Lee, Jaegul Choo
We present a novel Animation CelebHeads dataset (AnimeCeleb) to address an animation head reenactment.
1 code implementation • EMNLP 2021 • Jimin Hong, Taehee Kim, Hyesu Lim, Jaegul Choo
During the fine-tuning phase of transfer learning, the pretrained vocabulary remains unchanged, while model parameters are updated.
1 code implementation • 22 Oct 2021 • Jungsoo Lee, Jooyeol Yun, Sunghyun Park, Yonggyu Kim, Jaegul Choo
Despite the unprecedented improvement of face recognition, existing face recognition models still show considerably low performances in determining whether a pair of child and adult images belong to the same identity.
no code implementations • 18 Oct 2021 • Jeonghoon Park, Jimin Hong, Radhika Dua, Daehoon Gwak, Yixuan Li, Jaegul Choo, Edward Choi
Despite the impressive performance of deep networks in vision, language, and healthcare, unpredictable behaviors on samples from the distribution different than the training distribution cause severe problems in deployment.
no code implementations • 29 Sep 2021 • Daehoon Gwak, Gyubok Lee, Jaehoon Lee, Jaesik Choi, Jaegul Choo, Edward Choi
To address this, we introduce a new neural stochastic processes, Decoupled Kernel Neural Processes (DKNPs), which explicitly learn a separate mean and kernel function to directly model the covariance between output variables in a data-driven manner.
no code implementations • 29 Sep 2021 • Wonwoo Cho, Jaegul Choo
In this paper, we propose a novel distance-based BCR method suitable for OSR, which limits the feature space of known-class data in a class-wise manner and then makes background-class samples located far away from the limited feature space.
2 code implementations • ICLR 2022 • Taesung Kim, Jinhee Kim, Yunwon Tae, Cheonbok Park, Jang-Ho Choi, Jaegul Choo
The former normalizes the input to fix its distribution in terms of the mean and variance, while the latter returns the output to the original distribution.
no code implementations • ICCV 2021 • Eungyeup Kim, Jihyeon Lee, Jaegul Choo
Although previous approaches pre-define the type of dataset bias to prevent the network from learning it, recognizing the bias type in the real dataset is often prohibitive.
Ranked #3 on
Facial Attribute Classification
on bFFHQ
no code implementations • 17 Aug 2021 • Hojoon Lee, Dongyoon Hwang, Sunghwan Hong, Changyeon Kim, Seungryong Kim, Jaegul Choo
Successful sequential recommendation systems rely on accurately capturing the user's short-term and long-term interest.
1 code implementation • ICCV 2021 • Sanghun Jung, Jungsoo Lee, Daehoon Gwak, Sungha Choi, Jaegul Choo
However, the distribution of max logits of each predicted class is significantly different from each other, which degrades the performance of identifying unexpected objects in urban-scene segmentation.
Ranked #4 on
Anomaly Detection
on Lost and Found
1 code implementation • ACL 2021 • Nyoungwoo Lee, Suwon Shin, Jaegul Choo, Ho-Jin Choi, Sung-Hyun Myaeng
In multi-modal dialogue systems, it is important to allow the use of images as part of a multi-turn conversation.
1 code implementation • ICCV 2021 • Eungyeup Kim, Sanghyeon Lee, Jeonghoon Park, Somi Choi, Choonghyun Seo, Jaegul Choo
Deep neural networks for automatic image colorization often suffer from the color-bleeding artifact, a problematic color spreading near the boundaries between adjacent objects.
1 code implementation • NeurIPS 2021 • Jungsoo Lee, Eungyeup Kim, Juyoung Lee, Jihyeon Lee, Jaegul Choo
To this end, our method learns the disentangled representation of (1) the intrinsic attributes (i. e., those inherently defining a certain class) and (2) bias attributes (i. e., peripheral attributes causing the bias), from a large number of bias-aligned samples, the bias attributes of which have strong correlation with the target variable.
no code implementations • CVPR 2021 • Daejin Kim, Mohammad Azam Khan, Jaegul Choo
While the existing cycle-consistency loss ensures that the image can be translated back, our approach makes the model further preserve the attribute-irrelevant regions even in a single translation to another domain by using the Grad-CAM output computed from the discriminator.
no code implementations • 12 May 2021 • Hyunwook Lee, Cheonbok Park, Seungmin Jin, Hyeshin Chu, Jaegul Choo, Sungahn Ko
For example, it is difficult to figure out which models provide state-of-the-art performance, as recently proposed models have often been evaluated with different datasets and experiment environments.
no code implementations • NAACL 2021 • Kyeongpil Kang, Kyohoon Jin, Soyoung Yang, Sujin Jang, Jaegul Choo, Youngbin Kim
Understanding voluminous historical records provides clues on the past in various aspects, such as social and political issues and even natural science facts.
1 code implementation • CVPR 2021 • Seunghwan Choi, Sunghyun Park, Minsoo Lee, Jaegul Choo
The task of image-based virtual try-on aims to transfer a target clothing item onto the corresponding region of a person, which is commonly tackled by fitting the item to the desired body part and fusing the warped item with the person.
Ranked #5 on
Virtual Try-on
on VITON-HD
2 code implementations • CVPR 2021 • Sungha Choi, Sanghun Jung, Huiwon Yun, Joanne Kim, Seungryong Kim, Jaegul Choo
Enhancing the generalization capability of deep neural networks to unseen domains is crucial for safety-critical applications in the real world such as autonomous driving.
Ranked #6 on
Robust Object Detection
on DWD
no code implementations • 11 Feb 2021 • Taewoo Kim, Chaeyeon Chung, Sunghyun Park, Gyojung Gu, Keonmin Nam, Wonzo Choe, Jaesung Lee, Jaegul Choo
In response, we introduce a novel large-scale Korean hairstyle dataset, K-hairstyle, containing 500, 000 high-resolution images.
no code implementations • 1 Jan 2021 • Youngwoo Cho, Beomsoo Kim, Jaegul Choo
This paper considers neural networks as novel steganographic cover media, which we call stego networks, that can be used to hide one's secret messages.
no code implementations • 1 Jan 2021 • Junsoo Lee, Hojoon Lee, Inkyu Shin, Jaekyoung Bae, In So Kweon, Jaegul Choo
Learning visual representations using large-scale unlabelled images is a holy grail for most of computer vision tasks.
no code implementations • 1 Jan 2021 • Seohyun Back, Akhil Kedia, Sai Chetan Chinthakindi, Haejun Lee, Jaegul Choo
We evaluate our method against existing ones in terms of the quality of generated questions as well as the fine-tuned MRC model accuracy after training on the data synthetically generated by our method.
Ranked #3 on
Question Generation
on SQuAD1.1
(using extra training data)
no code implementations • 26 Nov 2020 • Jeonghoon Park, Kyungmin Jo, Daehoon Gwak, Jimin Hong, Jaegul Choo, Edward Choi
We evaluate the out-of-distribution (OOD) detection performance of self-supervised learning (SSL) techniques with a new evaluation framework.
Out-of-Distribution Detection
Out of Distribution (OOD) Detection
+1
1 code implementation • 19 Oct 2020 • Junwoo Park, Youngwoo Cho, Haneol Lee, Jaegul Choo, Edward Choi
Question Answering (QA) is a widely-used framework for developing and evaluating an intelligent machine.
no code implementations • ACL 2021 • Cheonbok Park, Yunwon Tae, Taehee Kim, Soyoung Yang, Mohammad Azam Khan, Eunjeong Park, Jaegul Choo
To address this issue, this paper presents a novel meta-learning algorithm for unsupervised neural machine translation (UNMT) that trains the model to adapt to another domain by utilizing only a small amount of training data.
1 code implementation • 16 Oct 2020 • Daehoon Gwak, Gyuhyeon Sim, Michael Poli, Stefano Massaroli, Jaegul Choo, Edward Choi
By interpreting the forward dynamics of the latent representation of neural networks as an ordinary differential equation, Neural Ordinary Differential Equation (Neural ODE) emerged as an effective framework for modeling a system dynamics in the continuous time domain.
1 code implementation • 16 Oct 2020 • Sunghyun Park, Kangyeol Kim, Junsoo Lee, Jaegul Choo, Joonseok Lee, Sookyung Kim, Edward Choi
Video generation models often operate under the assumption of fixed frame rates, which leads to suboptimal performance when it comes to handling flexible frame rates (e. g., increasing the frame rate of the more dynamic portion of the video as well as handling missing video frames).
no code implementations • 4 Sep 2020 • Heungseok Park, Yoonsoo Nam, Ji-Hoon Kim, Jaegul Choo
HyperTendril takes a novel approach to effectively steering hyperparameter optimization through an iterative, interactive tuning procedure that allows users to refine the search spaces and the configuration of the AutoML method based on their own insights from given results.
1 code implementation • arXiv.org 2020 • Seokwoo Jung, Sungha Choi, Mohammad Azam Khan, Jaegul Choo
This paper addresses the problem that pixel embedding in proposal-free instance segmentation based lane detection is difficult to optimize.
Ranked #11 on
Lane Detection
on TuSimple
no code implementations • CVPR 2020 • Junsoo Lee, Eungyeup Kim, Yunsung Lee, Dongjun Kim, Jaehyuk Chang, Jaegul Choo
However, it is difficult to prepare for a training data set that has a sufficient amount of semantically meaningful pairs of images as well as the ground truth for a colored image reflecting a given reference (e. g., coloring a sketch of an originally blue car given a reference green car).
no code implementations • ICLR 2020 • Seohyun Back, Sai Chetan Chinthakindi, Akhil Kedia, Haejun Lee, Jaegul Choo
Real-world question answering systems often retrieve potentially relevant documents to a given question through a keyword search, followed by a machine reading comprehension (MRC) step to find the exact answer from them.
1 code implementation • CVPR 2020 • Sungha Choi, Joanne T. Kim, Jaegul Choo
This paper exploits the intrinsic features of urban-scene images and proposes a general add-on module, called height-driven attention networks (HANet), for improving semantic segmentation for urban-scene images.
Ranked #17 on
Semantic Segmentation
on Cityscapes test
1 code implementation • CVPR 2020 • Hyojin Bahng, Sunghyo Chung, Seungjoo Yoo, Jaegul Choo
Despite remarkable success in unpaired image-to-image translation, existing systems still require a large amount of labeled images.
1 code implementation • 29 Nov 2019 • Cheonbok Park, Chunggi Lee, Hyojin Bahng, Yunwon Tae, Kihwan Kim, Seungmin Jin, Sungahn Ko, Jaegul Choo
Predicting road traffic speed is a challenging task due to different types of roads, abrupt speed change and spatial dependencies between roads; it requires the modeling of dynamically changing spatial dependencies among roads and temporal patterns over long input sequences.
no code implementations • 29 Nov 2019 • Wonwoong Cho, Kangyeol Kim, Eungyeup Kim, Hyunwoo J. Kim, Jaegul Choo
Disentangling content and style information of an image has played an important role in recent success in image translation.
no code implementations • IJCNLP 2019 • Fuxiang Chen, Seung-won Hwang, Jaegul Choo, Jung-Woo Ha, Sunghun Kim
Here we describe a new NL2pSQL task to generate pSQL codes from natural language questions on under-specified database issues, NL2pSQL.
3 code implementations • ICML 2020 • Hyojin Bahng, Sanghyuk Chun, Sangdoo Yun, Jaegul Choo, Seong Joon Oh
This tactic is feasible in many scenarios where it is much easier to define a set of biased representations than to define and quantify bias.
no code implementations • 25 Sep 2019 • Akhil Kedia, Sai Chetan Chinthakindi, Seohyun Back, Haejun Lee, Jaegul Choo
We evaluate the question generation capability of our method by comparing the BLEU score with existing methods and test our method by fine-tuning the MRC model on the downstream MRC data after training on synthetic data.
no code implementations • 25 Sep 2019 • Youngwoo Cho, Minwook Chang, Gerard Jounghyun Kim, Jaegul Choo
This paper proposes a novel generative model called PUGAN, which progressively synthesizes high-quality audio in a raw waveform.
no code implementations • 13 Sep 2019 • Cheonbok Park, Inyoup Na, Yongjang Jo, Sungbok Shin, Jaehyo Yoo, Bum Chul Kwon, Jian Zhao, Hyungjong Noh, Yeonsoo Lee, Jaegul Choo
Attention networks, a deep neural network architecture inspired by humans' attention mechanism, have seen significant success in image captioning, machine translation, and many other applications.
1 code implementation • 9 Jun 2019 • Seungjoo Yoo, Hyojin Bahng, Sunghyo Chung, Junsoo Lee, Jaehyuk Chang, Jaegul Choo
Despite recent advancements in deep learning-based automatic colorization, they are still limited when it comes to few-shot learning.
no code implementations • 9 Jun 2019 • Wonwoong Cho, Seunghwan Choi, Junwoo Park, David Keetae Park, Tao Qin, Jaegul Choo
First, those methods extract style from an entire exemplar which includes noisy information, which impedes a translation model from properly extracting the intended style of the exemplar.
1 code implementation • ICLR 2019 • Wonwoong Cho, Seunghwan Choi, Junwoo Park, David Keetae Park, Tao Qin, Jaegul Choo
Recently, image-to-image translation has seen a significant success.
no code implementations • 11 Feb 2019 • Sanghyeon Na, Seungjoo Yoo, Jaegul Choo
First, we use a content representation from the source domain conditioned on a style representation from the target domain.
2 code implementations • CVPR 2019 • Wonwoong Cho, Sungha Choi, David Keetae Park, Inkyu Shin, Jaegul Choo
However, applying this approach in image translation is computationally intensive and error-prone due to the expensive time complexity and its non-trivial backpropagation.
no code implementations • 13 Nov 2018 • Sungmin Kang, David Keetae Park, Jaehyuk Chang, Jaegul Choo
Questions convey information about the questioner, namely what one does not know.
no code implementations • EMNLP 2018 • Seohyun Back, Seunghak Yu, Sathish Reddy Indurthi, Jihie Kim, Jaegul Choo
Machine reading comprehension helps machines learn to utilize most of the human knowledge written in the form of text.
Ranked #36 on
Question Answering
on TriviaQA
(using extra training data)