Alert button
Picture for Sanghyuk Chun

Sanghyuk Chun

Alert button

StyleAugment: Learning Texture De-biased Representations by Style Augmentation without Pre-defined Textures

Aug 24, 2021
Sanghyuk Chun, Song Park

Figure 1 for StyleAugment: Learning Texture De-biased Representations by Style Augmentation without Pre-defined Textures
Figure 2 for StyleAugment: Learning Texture De-biased Representations by Style Augmentation without Pre-defined Textures
Figure 3 for StyleAugment: Learning Texture De-biased Representations by Style Augmentation without Pre-defined Textures
Figure 4 for StyleAugment: Learning Texture De-biased Representations by Style Augmentation without Pre-defined Textures
Viaarxiv icon

Neural Hybrid Automata: Learning Dynamics with Multiple Modes and Stochastic Transitions

Jun 08, 2021
Michael Poli, Stefano Massaroli, Luca Scimeca, Seong Joon Oh, Sanghyuk Chun, Atsushi Yamashita, Hajime Asama, Jinkyoo Park, Animesh Garg

Figure 1 for Neural Hybrid Automata: Learning Dynamics with Multiple Modes and Stochastic Transitions
Figure 2 for Neural Hybrid Automata: Learning Dynamics with Multiple Modes and Stochastic Transitions
Figure 3 for Neural Hybrid Automata: Learning Dynamics with Multiple Modes and Stochastic Transitions
Figure 4 for Neural Hybrid Automata: Learning Dynamics with Multiple Modes and Stochastic Transitions
Viaarxiv icon

Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts

Apr 02, 2021
Song Park, Sanghyuk Chun, Junbum Cha, Bado Lee, Hyunjung Shim

Figure 1 for Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts
Figure 2 for Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts
Figure 3 for Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts
Figure 4 for Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts
Viaarxiv icon

Rethinking Spatial Dimensions of Vision Transformers

Mar 30, 2021
Byeongho Heo, Sangdoo Yun, Dongyoon Han, Sanghyuk Chun, Junsuk Choe, Seong Joon Oh

Figure 1 for Rethinking Spatial Dimensions of Vision Transformers
Figure 2 for Rethinking Spatial Dimensions of Vision Transformers
Figure 3 for Rethinking Spatial Dimensions of Vision Transformers
Figure 4 for Rethinking Spatial Dimensions of Vision Transformers
Viaarxiv icon

Probabilistic Embeddings for Cross-Modal Retrieval

Jan 13, 2021
Sanghyuk Chun, Seong Joon Oh, Rafael Sampaio de Rezende, Yannis Kalantidis, Diane Larlus

Figure 1 for Probabilistic Embeddings for Cross-Modal Retrieval
Figure 2 for Probabilistic Embeddings for Cross-Modal Retrieval
Figure 3 for Probabilistic Embeddings for Cross-Modal Retrieval
Figure 4 for Probabilistic Embeddings for Cross-Modal Retrieval
Viaarxiv icon

Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels

Jan 13, 2021
Sangdoo Yun, Seong Joon Oh, Byeongho Heo, Dongyoon Han, Junsuk Choe, Sanghyuk Chun

Figure 1 for Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels
Figure 2 for Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels
Figure 3 for Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels
Figure 4 for Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels
Viaarxiv icon

Few-shot Font Generation with Localized Style Representations and Factorization

Sep 23, 2020
Song Park, Sanghyuk Chun, Junbum Cha, Bado Lee, Hyunjung Shim

Figure 1 for Few-shot Font Generation with Localized Style Representations and Factorization
Figure 2 for Few-shot Font Generation with Localized Style Representations and Factorization
Figure 3 for Few-shot Font Generation with Localized Style Representations and Factorization
Figure 4 for Few-shot Font Generation with Localized Style Representations and Factorization
Viaarxiv icon

Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets

Jul 08, 2020
Junsuk Choe, Seong Joon Oh, Sanghyuk Chun, Zeynep Akata, Hyunjung Shim

Figure 1 for Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets
Figure 2 for Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets
Figure 3 for Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets
Figure 4 for Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets
Viaarxiv icon

Slowing Down the Weight Norm Increase in Momentum-based Optimizers

Jun 15, 2020
Byeongho Heo, Sanghyuk Chun, Seong Joon Oh, Dongyoon Han, Sangdoo Yun, Youngjung Uh, Jung-Woo Ha

Figure 1 for Slowing Down the Weight Norm Increase in Momentum-based Optimizers
Figure 2 for Slowing Down the Weight Norm Increase in Momentum-based Optimizers
Figure 3 for Slowing Down the Weight Norm Increase in Momentum-based Optimizers
Figure 4 for Slowing Down the Weight Norm Increase in Momentum-based Optimizers
Viaarxiv icon

Few-shot Compositional Font Generation with Dual Memory

May 21, 2020
Junbum Cha, Sanghyuk Chun, Gayoung Lee, Bado Lee, Seonghyeon Kim, Hwalsuk Lee

Figure 1 for Few-shot Compositional Font Generation with Dual Memory
Figure 2 for Few-shot Compositional Font Generation with Dual Memory
Figure 3 for Few-shot Compositional Font Generation with Dual Memory
Figure 4 for Few-shot Compositional Font Generation with Dual Memory
Viaarxiv icon