Sungmin Eum, Ph.D.

I am a RESEARCH SCIENTIST at U.S. Army Research Laboratory. I obtained my Ph.D. from the University of Maryland, College Park in May 2017 under the supervision of Prof. David Doermann who is now the chair of the CSE Dept. at the University of Buffalo. My awesome committee members include Prof. Joseph JaJa, Prof. Larry Davis, Prof. Rama Chellappa, and Prof. Ramani Duraiswami. I acquired my B.S. and M.S. from the Yonsei University (Korea) in 2009 and 2011, respectively. During my Ph.D., I was fortunate to work as a research intern at Palo Alto Research Center (now part of SRI-Stanford Research Institute) and at the U.S. Army Research Laboratory.

I also serve as an ADJUNCT FACULTY in the ECE Department at the University of Maryland. I teach Digital Signal Processing (Fall) and Deep Learning for Computer Vision (Spring).

I do research in computer vision, machine learning, artificial intelligence, robotics, and natural language processing. I am particularly interested in creating models that can learn to leverage visual and/or language information to perform complex tasks with minimal supervision.

Email: ____ dot civ at army dot mil
    smeum at umd dot edu

News

• 10. 2024: Attending IROS 2024 in Abu Dhabi, UAE -- Will be presenting [oral] -- Two Teachers Are Better Than One: Leveraging Depth In Training Only For Unsupervised Obstacle Segmentation.
• 8. 2024: A paper under review SynPlay Dataset -- project page.
• 2. 2023: Attending AAAI 2023 in Washington DC -- Will be presenting at the Creative AI Across Modalities Workshop on 2/13.
• 1. 2023: A paper accepted at AAAI 2023 Workshop [Creative AI Across Modalities] -- SEE&TELL: Controllable Narrative Generation from Images
• 7. 2022: A paper accepted at ECCV -- Negative Samples are at Large: Leveraging Hard-distance Elastic Loss for Re-identification (code available!)
• 6. 2022: A paper accepted at Symposium on Combinatorial Search (SoCS) -- MA3: Model-Accuracy Aware Anytime Planning with Simulation Verification for Navigating Complex Terrains (Collaboration with CMU)
• 6. 2021: A paper accepted at IEEE Access -- Sketch-and-Fill Network for Semantic Segmentation
• 1. 2021: ME R-CNN featured in Army Top 10 Advances of 2020 (link)
• 1. 2021: A paper accepted at ICPR 2021 -- Semantics to Space(S2S): Embedding semantics into spatial space for zero-shot verb-object query inferencing
• 8. 2020: Received a new grant from ARL (GRIT Funding) for "Jarvis: Can You Tell Me What You See?"
• 6. 2020: A paper accepted at CVPR 2020 Workshops -- SomethingFinder: Localizing undefined regions using referring expressions
• 5. 2020: A paper accepted at ICASSP 2020 -- S-DOD-CNN: Doubly Injecting Spatially-Preserved Object Information for Event Recognition

Publications

SynPlay: Importing Real-world Diversity for a Synthetic Human Dataset
*Jinsub Yim, *Hyungtae Lee, *Sungmin Eum (* = equal contribution) , Yi-Ting Shen, Yan Zhang, Heesung Kwon, Shuvra S. Bhattacharyya
arXiv, 2024.
[paper] [Project Page]

SEE&TELL: Controllable Narrative Generation from Images
*Stephanie Lukin, *Sungmin Eum (* = equal contribution)
AAAI Workshop: Creative AI Across Modalities (AAAI-W), 2023.
[paper]

Negative Samples are at Large: Leveraging Hard-distance Elastic Loss for Re-identification
Hyungtae Lee, Sungmin Eum, Heesung Kwon
European Conference on Computer Vision (ECCV), 2022.
[paper]   [code]

MA3: Model-Accuracy Aware Anytime Planning with Simulation Verification for Navigating Complex Terrains
Manash Pratim Das, Damon M. Canover, Sungmin Eum, Heesung Kwon, Maxim Likhachev
Symposium on Combinatorial Search (SoCS), 2022.
[paper]

Sketch-and-Fill Network for Semantic Segmentation
Youngsaeng Jin, Sungmin Eum, David Han, Hanseok Ko
IEEE Access, 2021.
[paper]

Semantics to Space(S2S): Embedding semantics into spatial space for zero-shot verb-object query inferencing
Sungmin Eum, Heesung Kwon
International Conference on Pattern Recognition (ICPR), 2020.
[paper]


S-DOD-CNN: Doubly injecting spatially-preserved object information for event recognition
Hyungtae Lee, Sungmin Eum, Heesung Kwon
International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020.
[paper]

SomethingFinder: Localizing Undefined Regions Using Referring Expressions
Sungmin Eum, David Han, Gordon Briggs
IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2020.
[paper]

ME R-CNN: Multi-expert R-CNN for Object Detection
Hyungtae Lee, Sungmin Eum, Heesung Kwon
IEEE Transactions on Image Processing (TIP), 2019.
[paper]

A RUGD Dataset for Autonomous Navigation and Visual Perception in Unstructured Outdoor Environments
Maggie Wigness, Sungmin Eum, John G Rogers, David Han, Heesung Kwon
International Conference on Intelligent Robots and Systems (IROS), 2019.
[paper] [dataset]

Is Pretraining Necessary for hyperspectral image classification?
Hyungtae Lee, Sungmin Eum, Heesung Kwon
IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 2019.
[paper]

Planar content selection in images and videos using frontalness
Sungmin Eum, David Doermann
Pattern Recognition Letters (PRL), 2019.
[paper]

DOD-CNN: Doubly-injecting Object Information for Event Recognition
Hyungtae Lee, Sungmin Eum, Heesung Kwon
International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2019.
[paper]

Object and Text-guided Semantics for CNN-based Activity Recognition
Sungmin Eum, Christopher Reale, Heesung Kwon, Claire Bonial, Clare Voss
International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2019.
[paper]

Cross-domain CNN for hyperspectral image classification
Hyungtae Lee, Sungmin Eum, Heesung Kwon
IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 2018.
[paper]

Going deeper with CNN in malicious crowd event classification
Sungmin Eum, Hyungtae Lee, Heesung Kwon
SPIE Defense and Commercial Sensing (SPIE DCS), 2018.
[paper]

Exploitation of semantic keywords for malicious event classification
*Hyungtae Lee, *Sungmin Eum, *Joel Levis, Heesung Kwon, James Michaelis, Michael Kolodny
(* = equal contribution)
International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018.
[paper]

IOD-CNN: Integrating object detection networks for event recognition
*Sungmin Eum, *Hyungtae Lee, Heesung Kwon, David Doermann
(* = equal contribution)
International Conference on Image Processing (ICIP), 2017.
[paper]

Content selection using frontalness evaluation of multiple frames(Best Student Paper)
Sungmin Eum, David Doermann
International Conference on Pattern Recognition (ICPR), 2016.
[paper]


JH2R: Joint Homography Estimation for Highlight Removal
Sungmin Eum, Hyungtae Lee, David Doermann
British Machine Vision Conference (BMVC), 2015.
[paper]

Sharpness-aware document image mosaicing using graphcuts
Sungmin Eum, David Doermann
International Conference on Image Processing (ICIP), 2014.
[paper]

Enhancing Light Blob Detection for Intelligent Headlight Control Using Lane Detection
Sungmin Eum, Ho Gi Jung
IEEE Transactions on Intelligent Transportation Systems (T-ITS), 2013.
[paper]

Recognizability assessment of facial images for automated teller machine applications
Jae Kyu Suhr, Sungmin Eum, Ho Gi Jung, Gen Li, Gahyun Kim, Jaihie Kim
Pattern Recognition (PR), 2012.
[paper]

Face liveness detection based on texture and frequency analyses
Gahyun Kim, Sungmin Eum, Jae Kyu Suhr, Dong Ik Kim, Kang Ryoung Park, Jaihie Kim
International Conference on Biometrics (ICB), 2012.
[paper]

Face recognizability evaluation for ATM applications with exceptional occlusion handling
Sungmin Eum, Jae Kyu, Suhr, Jaihie Kim
IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2011.
[paper]

Professional Services

Editor

• Guest Editor- Remote Sensing (ISSN 2072-4292) Special Issue: Convolutional Neural Networks Applications in Remote Sensing II
• Guest Editor– Remote Sensing (ISSN 2072-4292) Special Issue: Computer Vision and and Deep Learning for Remote Sensing Applications

Reviewer | Program Committee

• IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2023
• International Conference on Computer Vision (ICCV) 2023
• European Conference on Computer Vision (ECCV) 2022, 2024
• Annual Conference on Neural Information Processing Systems (NEURIPS) 2020, 2022, 2024
• International Conference on Machine Learning (ICML) 2021, 2022, 2023, 2024
• International Conference on Learning Representations (ICLR) 2021, 2022, 2023
• AAAI Conference on Artificial Intelligence (AAAI) 2022, 2025
• International Conference on Intelligent Robots and Systems (IROS) 2019, 2022
• International Conference on Pattern Recognition (ICPR) 2016, 2022
• International Conference on Document Analysis and Recognition (ICDAR) 2019
• IEEE Transactions on Aerospace and Electronics Systems