Alert button
Picture for Nojun Kwak

Nojun Kwak

Alert button

Self-Evolutionary Optimization for Pareto Front Learning

Oct 07, 2021
Simyung Chang, KiYoon Yoo, Jiho Jang, Nojun Kwak

Figure 1 for Self-Evolutionary Optimization for Pareto Front Learning
Figure 2 for Self-Evolutionary Optimization for Pareto Front Learning
Figure 3 for Self-Evolutionary Optimization for Pareto Front Learning
Figure 4 for Self-Evolutionary Optimization for Pareto Front Learning
Viaarxiv icon

Few-Shot Object Detection by Attending to Per-Sample-Prototype

Sep 16, 2021
Hojun Lee, Myunggi Lee, Nojun Kwak

Figure 1 for Few-Shot Object Detection by Attending to Per-Sample-Prototype
Figure 2 for Few-Shot Object Detection by Attending to Per-Sample-Prototype
Figure 3 for Few-Shot Object Detection by Attending to Per-Sample-Prototype
Figure 4 for Few-Shot Object Detection by Attending to Per-Sample-Prototype
Viaarxiv icon

Dynamic Collective Intelligence Learning: Finding Efficient Sparse Model via Refined Gradients for Pruned Weights

Sep 10, 2021
Jangho Kim, Jayeon Yoo, Yeji Song, KiYoon Yoo, Nojun Kwak

Figure 1 for Dynamic Collective Intelligence Learning: Finding Efficient Sparse Model via Refined Gradients for Pruned Weights
Figure 2 for Dynamic Collective Intelligence Learning: Finding Efficient Sparse Model via Refined Gradients for Pruned Weights
Figure 3 for Dynamic Collective Intelligence Learning: Finding Efficient Sparse Model via Refined Gradients for Pruned Weights
Figure 4 for Dynamic Collective Intelligence Learning: Finding Efficient Sparse Model via Refined Gradients for Pruned Weights
Viaarxiv icon

Normalization Matters in Weakly Supervised Object Localization

Jul 28, 2021
Jeesoo Kim, Junsuk Choe, Sangdoo Yun, Nojun Kwak

Figure 1 for Normalization Matters in Weakly Supervised Object Localization
Figure 2 for Normalization Matters in Weakly Supervised Object Localization
Figure 3 for Normalization Matters in Weakly Supervised Object Localization
Figure 4 for Normalization Matters in Weakly Supervised Object Localization
Viaarxiv icon

PQK: Model Compression via Pruning, Quantization, and Knowledge Distillation

Jun 25, 2021
Jangho Kim, Simyung Chang, Nojun Kwak

Figure 1 for PQK: Model Compression via Pruning, Quantization, and Knowledge Distillation
Figure 2 for PQK: Model Compression via Pruning, Quantization, and Knowledge Distillation
Figure 3 for PQK: Model Compression via Pruning, Quantization, and Knowledge Distillation
Viaarxiv icon

LFI-CAM: Learning Feature Importance for Better Visual Explanation

May 03, 2021
Kwang Hee Lee, Chaewon Park, Junghyun Oh, Nojun Kwak

Figure 1 for LFI-CAM: Learning Feature Importance for Better Visual Explanation
Figure 2 for LFI-CAM: Learning Feature Importance for Better Visual Explanation
Figure 3 for LFI-CAM: Learning Feature Importance for Better Visual Explanation
Figure 4 for LFI-CAM: Learning Feature Importance for Better Visual Explanation
Viaarxiv icon

The U-Net based GLOW for Optical-Flow-free Video Interframe Generation

Apr 06, 2021
Saem Park, Donghoon Han, Nojun Kwak

Figure 1 for The U-Net based GLOW for Optical-Flow-free Video Interframe Generation
Figure 2 for The U-Net based GLOW for Optical-Flow-free Video Interframe Generation
Figure 3 for The U-Net based GLOW for Optical-Flow-free Video Interframe Generation
Figure 4 for The U-Net based GLOW for Optical-Flow-free Video Interframe Generation
Viaarxiv icon

Prototype-based Personalized Pruning

Mar 25, 2021
Jangho Kim, Simyung Chang, Sungrack Yun, Nojun Kwak

Figure 1 for Prototype-based Personalized Pruning
Figure 2 for Prototype-based Personalized Pruning
Figure 3 for Prototype-based Personalized Pruning
Figure 4 for Prototype-based Personalized Pruning
Viaarxiv icon

Maximizing Cosine Similarity Between Spatial Features for Unsupervised Domain Adaptation in Semantic Segmentation

Mar 18, 2021
Inseop Chung, Daesik Kim, Nojun Kwak

Figure 1 for Maximizing Cosine Similarity Between Spatial Features for Unsupervised Domain Adaptation in Semantic Segmentation
Figure 2 for Maximizing Cosine Similarity Between Spatial Features for Unsupervised Domain Adaptation in Semantic Segmentation
Figure 3 for Maximizing Cosine Similarity Between Spatial Features for Unsupervised Domain Adaptation in Semantic Segmentation
Figure 4 for Maximizing Cosine Similarity Between Spatial Features for Unsupervised Domain Adaptation in Semantic Segmentation
Viaarxiv icon

The Invertible U-Net for Optical-Flow-free Video Interframe Generation

Mar 17, 2021
Saem Park, Donghun Han, Nojun Kwak

Figure 1 for The Invertible U-Net for Optical-Flow-free Video Interframe Generation
Figure 2 for The Invertible U-Net for Optical-Flow-free Video Interframe Generation
Figure 3 for The Invertible U-Net for Optical-Flow-free Video Interframe Generation
Figure 4 for The Invertible U-Net for Optical-Flow-free Video Interframe Generation
Viaarxiv icon