no code implementations • 18 Mar 2024 • Seungbeom Woo, Geonwoo Baek, TaeHoon Kim, Jaemin Na, Joong-won Hwang, Wonjun Hwang
This framework dynamically cycles through multiple target domains, aligning each domain individually to restrain the biased alignment problem, and utilizes Fisher information to minimize the forgetting of knowledge from previous target domains.
1 code implementation • 14 Mar 2024 • Dinh Phat Do, TaeHoon Kim, Jaemin Na, Jiwon Kim, Keonho Lee, Kyunghwan Cho, Wonjun Hwang
However, there are limited studies on adapting from the visible to the thermal domain, because the domain gap between the visible and thermal domains is much larger than expected, and traditional domain adaptation can not successfully facilitate learning in this situation.
no code implementations • 11 Mar 2024 • Jongwook Choi, TaeHoon Kim, Yonghyun Jeong, Seungryul Baek, Jongwon Choi
This paper presents a new approach for the detection of fake videos, based on the analysis of style latent vectors and their abnormal behavior in temporal changes in the generated videos.
no code implementations • 5 Sep 2023 • TaeHoon Kim, Pyunghwan Ahn, Sangyun Kim, Sihaeng Lee, Mark Marsden, Alessandra Sala, Seung Hwan Kim, Bohyung Han, Kyoung Mu Lee, Honglak Lee, Kyounghoon Bae, Xiangyu Wu, Yi Gao, Hailiang Zhang, Yang Yang, Weili Guo, Jianfeng Lu, Youngtaek Oh, Jae Won Cho, Dong-Jin Kim, In So Kweon, Junmo Kim, Wooyoung Kang, Won Young Jhoo, Byungseok Roh, Jonghwan Mun, Solgil Oh, Kenan Emir Ak, Gwang-Gook Lee, Yan Xu, Mingwei Shen, Kyomin Hwang, Wonsik Shin, Kamin Lee, Wonhark Park, Dongkwan Lee, Nojun Kwak, Yujin Wang, Yimu Wang, Tiancheng Gu, Xingchang Lv, Mingmao Sun
In this report, we introduce NICE (New frontiers for zero-shot Image Captioning Evaluation) project and share the results and outcomes of 2023 challenge.
1 code implementation • 19 May 2023 • Haram Choi, Cheolwoong Na, Jihyeon Oh, Seungjae Lee, Jinseop Kim, Subeen Choe, Jeongmin Lee, TaeHoon Kim, Jihoon Yang
To address these problems, we propose a lightweight IR network, Reciprocal Attention Mixing Transformer (RAMiT).
no code implementations • 4 Apr 2023 • TaeHoon Kim, Bohyung Han
We propose a novel domain generalization technique, referred to as Randomized Adversarial Style Perturbation (RASP), which is motivated by the observation that the characteristics of each domain are captured by the feature statistics corresponding to style.
no code implementations • 4 Apr 2023 • TaeHoon Kim, Jaeyoo Park, Bohyung Han
The proposed approach has a unique perspective to utilize the previous knowledge in class incremental learning since it augments features of arbitrary target classes using examples in other classes via adversarial attacks on a previously learned classifier.
1 code implementation • 13 Nov 2022 • TaeHoon Kim, Mark Marsden, Pyunghwan Ahn, Sangyun Kim, Sihaeng Lee, Alessandra Sala, Seung Hwan Kim
However, we find that large-scale bidirectional training between image and text enables zero-shot image captioning.
1 code implementation • 15 Mar 2022 • Jinsu Yoo, TaeHoon Kim, Sihaeng Lee, Seung Hwan Kim, Honglak Lee, Tae Hyun Kim
Recent transformer-based super-resolution (SR) methods have achieved promising results against conventional CNN-based methods.
1 code implementation • CVPR 2022 • TaeHoon Kim, Gwangmo Song, Sihaeng Lee, Sangyun Kim, Yewon Seo, Soonyoung Lee, Seung Hwan Kim, Honglak Lee, Kyunghoon Bae
Unlike other models, BiART can distinguish between image (or text) as a conditional reference and a generation target.
Ranked #1 on Image Reconstruction on ImageNet 256x256
no code implementations • 19 Nov 2021 • Alexandra Vioni, Myrsini Christidou, Nikolaos Ellinas, Georgios Vamvoukakis, Panos Kakoulidis, TaeHoon Kim, June Sig Sung, Hyoungmin Park, Aimilios Chalamandaris, Pirros Tsiakoulis
This paper presents a method for controlling the prosody at the phoneme level in an autoregressive attention-based text-to-speech system.
1 code implementation • 17 Jun 2020 • Taehoon Kim, Youngjoon Yoo, Jihoon Yang
In this paper, we present a new network architecture search (NAS) procedure to find a network that guarantees both full-precision (FLOAT32) and quantized (INT8) performances.
1 code implementation • 20 May 2018 • Taehoon Kim, Jihoon Yang
Seq2CNN is trained end-to-end to classify various-length texts without preprocessing inputs into fixed length.
Ranked #10 on Text Classification on Yahoo! Answers