no code implementations • 26 Apr 2024 • Wonjae Kim, Sanghyuk Chun, Taekyung Kim, Dongyoon Han, Sangdoo Yun
In an era where the volume of data drives the effectiveness of self-supervised learning, the specificity and clarity of data semantics play a crucial role in model training.
no code implementations • 15 Apr 2024 • Minji Kim, Dongyoon Han, Taekyung Kim, Bohyung Han
We propose Temporal Contextualization (TC), a novel layer-wise temporal information infusion mechanism for video that extracts core information from each frame, interconnects relevant information across the video to summarize into context tokens, and ultimately leverages the context tokens during the feature encoding process.
no code implementations • 30 Dec 2023 • Taekyung Kim, Dongyoon Han, Byeongho Heo
Masked Image Modeling (MIM) arises as a promising option for Vision Transformers among various self-supervised learning (SSL) methods.
no code implementations • 20 Oct 2023 • Taekyung Kim, Sanghyuk Chun, Byeongho Heo, Dongyoon Han
MIMs such as Masked Autoencoder (MAE) learn strong representations by randomly masking input tokens for the encoder to process, with the decoder reconstructing the masked tokens to the input.
no code implementations • 26 Jul 2023 • Junwon Seo, Taekyung Kim, Seongyong Ahn, Kiho Kwak
To conduct a comprehensive evaluation, we collect driving data from various terrains and demonstrate that our method can obtain a global model that minimizes uncertainty.
no code implementations • 26 Jun 2023 • Junwon Seo, Jungwi Mun, Taekyung Kim
We train a vehicle dynamics model that can quantify the epistemic uncertainty of the model to perform active exploration, resulting in the efficient collection of training data and effective avoidance of uncertain state-action spaces.
1 code implementation • 20 Jun 2023 • Byeongho Heo, Taekyung Kim, Sangdoo Yun, Dongyoon Han
In this paper, we propose a novel way to involve masking augmentations dubbed Masked Sub-model (MaskSub).
no code implementations • 20 May 2023 • Taekyung Kim, Jungwi Mun, Junwon Seo, Beomsu Kim, Seongil Hong
Active exploration, in which a robot directs itself to states that yield the highest information gain, is essential for efficient data collection and minimizing human supervision.
1 code implementation • 1 May 2023 • Namuk Park, Wonjae Kim, Byeongho Heo, Taekyung Kim, Sangdoo Yun
We present a comparative study on how and why contrastive learning (CL) and masked image modeling (MIM) differ in their representations and in their performance of downstream tasks.
no code implementations • 1 May 2023 • Hojin Lee, Taekyung Kim, Jungwi Mun, Wonsuk Lee
High-speed autonomous driving in off-road environments has immense potential for various applications, but it also presents challenges due to the complexity of vehicle-terrain interactions.
no code implementations • 11 Apr 2023 • Soohyun Kim, Junho Kim, Taekyung Kim, Hwan Heo, Seungryong Kim, Jiyoung Lee, Jin-Hwa Kim
This task is difficult due to the geometric distortion of panoramic images and the lack of a panoramic image dataset with diverse conditions, like weather or time.
no code implementations • 3 Feb 2023 • Hwan Heo, Taekyung Kim, Jiyoung Lee, Jaewon Lee, Soohyun Kim, Hyunwoo J. Kim, Jin-Hwa Kim
Multi-resolution hash encoding has recently been proposed to reduce the computational cost of neural renderings, such as NeRF.
no code implementations • 21 Nov 2022 • Jihwan Bae, Junwon Seo, Taekyung Kim, Hae-Gon Jeon, Kiho Kwak, Inwook Shim
To mitigate the uncertainty, we introduce a deep metric learning-based method to incorporate unlabeled data with a few positive and negative prototypes in order to leverage the uncertainty, which jointly learns using semantic segmentation and traversability regression.
no code implementations • 14 Sep 2022 • Junwon Seo, Taekyung Kim, Kiho Kwak, Jihong Min, Inwook Shim
By integrating our framework with a model predictive controller, we demonstrate that estimated traversability results in effective navigation that enables distinct maneuvers based on the driving characteristics of the vehicles.
no code implementations • 16 Jul 2022 • Taekyung Kim, Hojin Lee, Wonsuk Lee
Non-holonomic vehicle motion has been studied extensively using physics-based models.
no code implementations • 8 Nov 2021 • Byeongjun Park, Taekyung Kim, Hyojun Go, Changick Kim
In this paper, we propose residual guidance loss that enables the depth estimation network to embed the discriminative feature by transferring the discriminability of auto-encoded features.
no code implementations • 1 Jan 2021 • Taekyung Kim, Changick Kim
We propose a photometric consistency loss, which directly enforces the geometrically consistent style texture across the view, and a stroke consistency loss, which matches the characteristics and directions of the brushstrokes by aligning the local patches of the corresponding pixels before minimizing feature deviation.
no code implementations • ICCV 2021 • Taekyung Kim, Jaehoon Choi, Seokeon Choi, Dongki Jung, Changick Kim
We generate the spare ground truth of the DTU dataset for evaluation and extensive experiments verify that our SGT-MVSNet outperforms the state-of-the-art MVS methods on the sparse ground truth setting.
1 code implementation • CVPR 2021 • Seokeon Choi, Taekyung Kim, Minki Jeong, Hyoungseob Park, Changick Kim
To this end, we combine learnable batch-instance normalization layers with meta-learning and investigate the challenging cases caused by both batch and instance normalization layers.
2 code implementations • ECCV 2020 • Taekyung Kim, Changick Kim
Finally, the exploration scheme locally aligns features in a class-wise manner complementary to the attraction scheme by selectively aligning unlabeled target features complementary to the perturbation scheme.
Semi-supervised Domain Adaptation Unsupervised Domain Adaptation
1 code implementation • CVPR 2020 • Seokeon Choi, Sumin Lee, Youngeun Kim, Taekyung Kim, Changick Kim
To implement our approach, we introduce an ID-preserving person image generation network and a hierarchical feature learning module.
no code implementations • 2 Oct 2019 • Youngeun Kim, Seunghyeon Kim, Taekyung Kim, Changick Kim
Note that each binary image consists of background and regions belonging to a class.
no code implementations • 29 Sep 2019 • Youngeun Kim, Seokeon Choi, Taekyung Kim, Sumin Lee, Changick Kim
Since the cost of labeling increases dramatically as the number of cameras increases, it is difficult to apply the re-identification algorithm to a large camera network.
no code implementations • 29 Sep 2019 • Youngeun Kim, Seokeon Choi, Hankyeol Lee, Taekyung Kim, Changick Kim
In this paper, we introduce a self-supervised approach for video object segmentation without human labeled data. Specifically, we present Robust Pixel-level Matching Net-works (RPM-Net), a novel deep architecture that matches pixels between adjacent frames, using only color information from unlabeled videos for training.
no code implementations • ICCV 2019 • Jaehoon Choi, Taekyung Kim, Changick Kim
Unsupervised domain adaptation seeks to adapt the model trained on the source domain to the target domain.
no code implementations • ICCV 2019 • Seunghyeon Kim, Jaehoon Choi, Taekyung Kim, Changick Kim
Experimental results show that our approach effectively improves the performance of the one-stage object detection in unsupervised domain adaptation setting.
no code implementations • 1 Aug 2019 • Jaehoon Choi, Minki Jeong, Taekyung Kim, Changick Kim
To learn target discriminative representations, using pseudo-labels is a simple yet effective approach for unsupervised domain adaptation.
no code implementations • CVPR 2019 • Taekyung Kim, Minki Jeong, Seunghyeon Kim, Seokeon Choi, Changick Kim
We construct a structured domain adaptation framework for our learning paradigm and introduce a practical way of DD for implementation.