Picture for Bumsub Ham

Bumsub Ham

ALIFE: Adaptive Logit Regularizer and Feature Replay for Incremental Semantic Segmentation

Oct 13, 2022
Youngmin Oh, Donghyeon Baek, Bumsub Ham

Figure 1 for ALIFE: Adaptive Logit Regularizer and Feature Replay for Incremental Semantic Segmentation

Figure 2 for ALIFE: Adaptive Logit Regularizer and Feature Replay for Incremental Semantic Segmentation

Figure 3 for ALIFE: Adaptive Logit Regularizer and Feature Replay for Incremental Semantic Segmentation

Figure 4 for ALIFE: Adaptive Logit Regularizer and Feature Replay for Incremental Semantic Segmentation

* Accepted to NeurIPS 2022

Access Paper or Ask Questions

Decomposed Knowledge Distillation for Class-Incremental Semantic Segmentation

Oct 12, 2022
Donghyeon Baek, Youngmin Oh, Sanghoon Lee, Junghyup Lee, Bumsub Ham

Figure 1 for Decomposed Knowledge Distillation for Class-Incremental Semantic Segmentation

Figure 2 for Decomposed Knowledge Distillation for Class-Incremental Semantic Segmentation

Figure 3 for Decomposed Knowledge Distillation for Class-Incremental Semantic Segmentation

Figure 4 for Decomposed Knowledge Distillation for Class-Incremental Semantic Segmentation

* Accepted to NeurIPS 2022

Access Paper or Ask Questions

Bi-directional Contrastive Learning for Domain Adaptive Semantic Segmentation

Jul 22, 2022
Geon Lee, Chanho Eom, Wonkyung Lee, Hyekang Park, Bumsub Ham

Figure 1 for Bi-directional Contrastive Learning for Domain Adaptive Semantic Segmentation

Figure 2 for Bi-directional Contrastive Learning for Domain Adaptive Semantic Segmentation

Figure 3 for Bi-directional Contrastive Learning for Domain Adaptive Semantic Segmentation

Figure 4 for Bi-directional Contrastive Learning for Domain Adaptive Semantic Segmentation

* Accepted to ECCV 2022

Access Paper or Ask Questions

OIMNet++: Prototypical Normalization and Localization-aware Learning for Person Search

Jul 21, 2022
Sanghoon Lee, Youngmin Oh, Donghyeon Baek, Junghyup Lee, Bumsub Ham

Figure 1 for OIMNet++: Prototypical Normalization and Localization-aware Learning for Person Search

Figure 2 for OIMNet++: Prototypical Normalization and Localization-aware Learning for Person Search

Figure 3 for OIMNet++: Prototypical Normalization and Localization-aware Learning for Person Search

Figure 4 for OIMNet++: Prototypical Normalization and Localization-aware Learning for Person Search

* Accepted to ECCV 2022

Access Paper or Ask Questions

Video-based Person Re-identification with Spatial and Temporal Memory Networks

Aug 20, 2021
Chanho Eom, Geon Lee, Junghyup Lee, Bumsub Ham

Figure 1 for Video-based Person Re-identification with Spatial and Temporal Memory Networks

Figure 2 for Video-based Person Re-identification with Spatial and Temporal Memory Networks

Figure 3 for Video-based Person Re-identification with Spatial and Temporal Memory Networks

Figure 4 for Video-based Person Re-identification with Spatial and Temporal Memory Networks

* International Conference on Computer Vision (ICCV) 2021

Access Paper or Ask Questions

Distance-aware Quantization

Aug 16, 2021
Dohyung kim, Junghyup Lee, Bumsub Ham

Figure 1 for Distance-aware Quantization

Figure 2 for Distance-aware Quantization

Figure 3 for Distance-aware Quantization

Figure 4 for Distance-aware Quantization

* ICCV2021

Access Paper or Ask Questions

Exploiting a Joint Embedding Space for Generalized Zero-Shot Semantic Segmentation

Aug 14, 2021
Donghyeon Baek, Youngmin Oh, Bumsub Ham

Figure 1 for Exploiting a Joint Embedding Space for Generalized Zero-Shot Semantic Segmentation

Figure 2 for Exploiting a Joint Embedding Space for Generalized Zero-Shot Semantic Segmentation

Figure 3 for Exploiting a Joint Embedding Space for Generalized Zero-Shot Semantic Segmentation

Figure 4 for Exploiting a Joint Embedding Space for Generalized Zero-Shot Semantic Segmentation

* Accepted to ICCV 2021

Access Paper or Ask Questions

Background-Aware Pooling and Noise-Aware Loss for Weakly-Supervised Semantic Segmentation

Apr 02, 2021
Youngmin Oh, Beomjun Kim, Bumsub Ham

Figure 1 for Background-Aware Pooling and Noise-Aware Loss for Weakly-Supervised Semantic Segmentation

Figure 2 for Background-Aware Pooling and Noise-Aware Loss for Weakly-Supervised Semantic Segmentation

Figure 3 for Background-Aware Pooling and Noise-Aware Loss for Weakly-Supervised Semantic Segmentation

Figure 4 for Background-Aware Pooling and Noise-Aware Loss for Weakly-Supervised Semantic Segmentation

* Accepted to CVPR 2021

Access Paper or Ask Questions

Network Quantization with Element-wise Gradient Scaling

Apr 02, 2021
Junghyup Lee, Dohyung Kim, Bumsub Ham

Figure 1 for Network Quantization with Element-wise Gradient Scaling

Figure 2 for Network Quantization with Element-wise Gradient Scaling

Figure 3 for Network Quantization with Element-wise Gradient Scaling

Figure 4 for Network Quantization with Element-wise Gradient Scaling

* Accepted to CVPR 2021

Access Paper or Ask Questions