Byeongho Heo's Hompage

[ Google scholar ] [ CV ] [linkedin] [github]

Ph.D. Byeongho Heo

Research scientist

NAVER AI LAB

Visiting assistant professor

AI institute, Seoul National University

Email : ho6946 [at] gmail.com, bh.heo [at] navercorp.com, bhheo [at] snu.ac.kr

Research interest: vision transformer, knowledge distillation, network compression, network architecture

Publications


Conferences

Joint Global and Local Hierarchical Priors for Learned Image Compression

Jun-Hyuk Kim, Byeongho Heo, Jong-Seok Lee

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022

[arxiv]

The Majority Can Help The Minority: Context-rich Minority Oversampling for Long-tailed Classification

Seulki Park, Youngkyu Hong, Byeongho Heo, Sangdoo Yun, Jin Young Choi

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022

[arxiv]

Learning Features with Parameter-Free Layers

Dongyoon Han, YoungJoon Yoo, Beomyoung Kim, Byeongho Heo

International Conference on Learning Representations (ICLR), 2022

[pdf] [arxiv]

ViDT: An Efficient and Effective Fully Transformer-based Object Detector

Hwanjun Song, Deqing Sun, Sanghyuk Chun, Varun Jampani, Dongyoon Han, Byeongho Heo, Wonjae Kim, Ming-Hsuan Yang

International Conference on Learning Representations (ICLR), 2022

[arxiv] [code]

Rethinking Spatial Dimensions of Vision Transformers

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

IEEE International Conference on Computer Vision (ICCV), 2021

[arxiv] [code]

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

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

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021

[arxiv] [code]

Rethinking Channel Dimensions for Efficient Model Design

Dongyoon Han, Sangdoo Yun, Byeongho Heo, YoungJoon Yoo

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021

[arxiv] [code]


AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights

Byeongho Heo*, Sanghyuk Chun*, Seong Joon Oh, Dongyoon Han, Sangdoo Yun, Gyuwan Kim, Youngjung Uh, Jung-Woo Ha

International Conference on Learning Representations (ICLR), 2021

[arxiv] [project page] [code]


Show, Attend and Distill: Knowledge Distillation via Attention-Based Feature Matching

Mingi Ji, Byeongho Heo, Sungrae Park

AAAI Conference on Artificial Intelligence (AAAI), 2021

[arxiv] [code]

VideoMix: Rethinking Data Augmentation for Video Classification

Sangdoo Yun , Seong Joon Oh , Byeongho Heo , Dongyoon Han , Jinhyung Kim

Arxiv, 2020

[arxiv]

A Comprehensive Overhaul of Feature Distillation

Byeongho Heo, Jeesoo Kim, Sangdoo Yun, Hyojin Park, Nojun Kwak, and Jin Young Choi

IEEE International Conference on Computer Vision (ICCV), 2019

[arxiv] [project page] [code] [slides] [poster]

Knowledge Transfer via Distillation of Activation Boundaries Formed by Hidden Neurons

Byeongho Heo, Minsik Lee, Sangdoo Yun, and Jin Young Choi

AAAI Conference on Artificial Intelligence (AAAI), 2019, oral

[pdf] [code] [slides] [poster]

Knowledge Distillation with Adversarial Samples Supporting Decision Boundary

Byeongho Heo*, Minsik Lee*, Sangdoo Yun, and Jin Young Choi

AAAI Conference on Artificial Intelligence (AAAI), 2019

[pdf] [code] [slides] [spotlight] [poster]

Backbone Can Not be Trained at Once: Rolling Back to Pre-trained Network for Person Re-identification

Youngmin Ro, Jongwon Choi, Dae Ung Jo, Byeongho Heo, Jongin Lim, and Jin Young Choi

AAAI Conference on Artificial Intelligence (AAAI), 2019

[pdf] [code]

Appearance and motion based deep learning architecture for moving object detection in moving camera

Byeongho Heo, Kimin Yun, and Jin Young Choi

IEEE International Conference on Image Processing (ICIP), 2017, oral

[pdf]

Deep Learning Architecture for Pedestrian 3-D Localization and Tracking Using Multiple Cameras

Kikyung Kim, Byeongho Heo, Moonsub Byeon, and Jin Young Choi

IEEE International Conference on Image Processing (ICIP), 2017, oral

[pdf]

Weighted Pooling Based on Visual Saliency for Image Classification

Byeongho Heo, Hawook Jeong, Jiyun Kim, Sang-Il Choi, and Jin Young Choi

International Symposium on Visual Computing (ISVC), 2014, oral

[pdf]

Journals

Detecting and Removing Text in the Wild

Junho Cho, Sangdoo Yun, Dongyoon Han, Byeongho Heo, Jin Young Choi

IEEE Access, 2021

[paper]

Motion-aware ensemble of three-mode trackers for unmanned aerial vehicles

Kyuewang Lee, Hyung Jin Chang, Jongwon Choi, Byeongho Heo, Aleš Leonardis, Jin Young Choi

Machine Vision and Applications, 2021

[paper]


Rollback Ensemble with Multiple Local Minima in Fine-tuning Deep Learning Networks

Youngmin Ro, Jongwon Choi, Byeongho Heo, Jin Young Choi

IEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2021.02

[paper]

Filter Pruning and Re-Initialization via Latent Space Clustering

Seunghyun Lee, Byeongho Heo, Jung-Woo Ha, Byung Cheol Song

IEEE Access, 2020

[paper]

Pose Transforming Network: Learning to Disentangle Human Posture in Variational Auto-encoded Latent Space

Jongin Lim, Youngjoon Yoo, Byeongho Heo, and Jin Young Choi

Pattern Recognition Letters (PRL), September 2018

[pdf]

Research interest

    • Deep learning

    • Knowledge distillation

    • Network compression

    • Image classification, object detection