no code implementations • 25 Oct 2023 • Christopher Maxey, Jaehoon Choi, Hyungtae Lee, Dinesh Manocha, Heesung Kwon
Using various synthetic renderers in conjunction with perception models is prevalent to create synthetic data to augment the learning in the ground-based imaging domain.
1 code implementation • 7 Jul 2023 • Zahid Hasan, Abu Zaher Md Faridee, Masud Ahmed, Sanjay Purushotham, Heesung Kwon, Hyungtae Lee, Nirmalya Roy
As an alternative to the traditional pseudo-labeling-based approaches, we leverage the connection between the data sampling and the provided multinoulli (categorical) distribution of novel classes.
1 code implementation • 14 Apr 2023 • Zahid Hasan, Masud Ahmed, Abu Zaher Md Faridee, Sanjay Purushotham, Heesung Kwon, Hyungtae Lee, Nirmalya Roy
During our experiments with UCF101 and multi-view action dataset, NEV-NCD achieves ~ 83% classification accuracy in test instances of labeled data.
no code implementations • CVPR 2023 • Yi-Ting Shen, Hyungtae Lee, Heesung Kwon, Shuvra Shikhar Bhattacharyya
To effectively interrogate UAV-based images for detecting objects of interest, such as humans, it is essential to acquire large-scale UAV-based datasets that include human instances with various poses captured from widely varying viewing angles.
no code implementations • 31 Aug 2022 • Yi-Ting Shen, Yaesop Lee, Heesung Kwon, Damon M. Conover, Shuvra S. Bhattacharyya, Nikolas Vale, Joshua D. Gray, G. Jeremy Leong, Kenneth Evensen, Frank Skirlo
Learning to detect objects, such as humans, in imagery captured by an unmanned aerial vehicle (UAV) usually suffers from tremendous variations caused by the UAV's position towards the objects.
no code implementations • 20 Jul 2022 • Hyungtae Lee, Sungmin Eum, Heesung Kwon
We present a Momentum Re-identification (MoReID) framework that can leverage a very large number of negative samples in training for general re-identification task.
no code implementations • 13 Jun 2022 • Priya Narayanan, Xin Hu, Zhenyu Wu, Matthew D Thielke, John G Rogers, Andre V Harrison, John A D'Agostino, James D Brown, Long P Quang, James R Uplinger, Heesung Kwon, Zhangyang Wang
The full dataset presented in this paper, including the ground truth object classification bounding boxes and haze density measurements, is provided for the community to evaluate their algorithms at: https://a2i2-archangel. vision.
no code implementations • 7 Apr 2022 • Hyungtae Lee, Sungmin Eum, Heesung Kwon
In addition, we have verified that our approach effectively reduces the overfitting issue, enabling us to deepen the model up to 13 layers (from 9) without compromising the accuracy.
no code implementations • 6 Apr 2022 • Hyungtae Lee, Heesung Kwon
In this paper, we propose a novel and highly practical score-level fusion approach called dynamic belief fusion (DBF) that directly integrates inference scores of individual detections from multiple object detection methods.
no code implementations • 8 Feb 2022 • Hyungtae Lee, Heesung Kwon
Recently, self-supervised learning has attracted attention due to its remarkable ability to acquire meaningful representations for classification tasks without using semantic labels.
no code implementations • 17 Jan 2022 • Eung-Joo Lee, Damon M. Conover, Shuvra S. Bhattacharyyaa, Heesung Kwon, Jason Hill, Kenneth Evensen
Using the synthetic datasets, we analyze detection accuracy in different imaging conditions as a function of the above parameters.
no code implementations • ICLR 2019 • Swati Rallapalli, Liang Ma, Mudhakar Srivatsa, Ananthram Swami, Heesung Kwon, Graham Bent, Christopher Simpkin
Effectively capturing graph node sequences in the form of vector embeddings is critical to many applications.
2 code implementations • ICCV 2019 • Zhen-Yu Wu, Karthik Suresh, Priya Narayanan, Hongyu Xu, Heesung Kwon, Zhangyang Wang
Object detection from images captured by Unmanned Aerial Vehicles (UAVs) is becoming increasingly useful.
no code implementations • 13 Jun 2019 • Sungmin Eum, Heesung Kwon
Our approach is powered by Semantics-to-Space (S2S) architecture where semantics derived from the residing objects are embedded into a spatial space of the visual stream.
no code implementations • 14 Feb 2019 • Colin Samplawski, Heesung Kwon, Erik Learned-Miller, Benjamin M. Marlin
Zero-shot learning (ZSL) is one of the most extreme forms of learning from scarce labeled data.
no code implementations • 11 Feb 2019 • Hyungtae Lee, Sungmin Eum, Heesung Kwon
We present a novel event recognition approach called Spatially-preserved Doubly-injected Object Detection CNN (S-DOD-CNN), which incorporates the spatially preserved object detection information in both a direct and an indirect way.
no code implementations • 24 Jan 2019 • Hyungtae Lee, Sungmin Eum, Heesung Kwon
To answer the first question, we have devised an approach that pre-trains a network on multiple source datasets that differ in their hyperspectral characteristics and fine-tunes on a target dataset.
no code implementations • 13 Dec 2018 • Hyungtae Lee, Heesung Kwon, Wonkook Kim
To overcome the problem of the data scarcity and lack of hard examples in training, we introduce a two-step hard example generation (HEG) approach that first generates hard example candidates and then mines actual hard examples.
no code implementations • 7 Nov 2018 • Hyungtae Lee, Sungmin Eum, Heesung Kwon
Recognizing an event in an image can be enhanced by detecting relevant objects in two ways: 1) indirectly utilizing object detection information within the unified architecture or 2) directly making use of the object detection output results.
no code implementations • COLING 2018 • Christopher Reale, Claire Bonial, Heesung Kwon, Clare Voss
We propose a method to improve human activity recognition in video by leveraging semantic information about the target activities from an expert-defined linguistic resource, VerbNet.
no code implementations • 4 May 2018 • Sungmin Eum, Christopher Reale, Heesung Kwon, Claire Bonial, Clare Voss
We further improve upon the multitask learning approach by exploiting a text-guided semantic space to select the most relevant objects with respect to the target activities.
no code implementations • 31 Jan 2018 • Hyungtae Lee, Sungmin Eum, Heesung Kwon
To cope with this problem, we propose a novel cross-domain CNN containing the shared parameters which can co-learn across multiple hyperspectral datasets.
no code implementations • 4 Apr 2017 • Hyungtae Lee, Sungmin Eum, Heesung Kwon
To address this problem, we introduce a practical training strategy which is tailored to optimize ME, EAN, and the shared network in an end-to-end fashion.
no code implementations • 21 Mar 2017 • Sungmin Eum, Hyungtae Lee, Heesung Kwon, David Doermann
Many previous methods have showed the importance of considering semantically relevant objects for performing event recognition, yet none of the methods have exploited the power of deep convolutional neural networks to directly integrate relevant object information into a unified network.
no code implementations • 21 Oct 2016 • Yilun Cao, Hyungtae Lee, Heesung Kwon
The prior knowledge is then fused with the decisions of object detection to improve detection accuracy by mitigating false positives of an object detector that are strongly contradicted with the prior knowledge.
no code implementations • 21 Oct 2016 • Hyungtae Lee, Sungmin Eum, Joel Levis, Heesung Kwon, James Michaelis, Michael Kolodny
Learning an event classifier is challenging when the scenes are semantically different but visually similar.
no code implementations • 12 Apr 2016 • Hyungtae Lee, Heesung Kwon, Ryan M. Robinson, William D. Nothwang
A novel template matching algorithm that can incorporate the concept of deformable parts, is presented in this paper.
2 code implementations • 12 Apr 2016 • Hyungtae Lee, Heesung Kwon
The initial spatial and spectral feature maps obtained from the multi-scale filter bank are then combined together to form a joint spatio-spectral feature map.
no code implementations • 12 Apr 2016 • Hyungtae Lee, Heesung Kwon, Archith J. Bency, William D. Nothwang
Object localization is an important task in computer vision but requires a large amount of computational power due mainly to an exhaustive multiscale search on the input image.
no code implementations • 1 Mar 2016 • Archith J. Bency, Heesung Kwon, Hyungtae Lee, S. Karthikeyan, B. S. Manjunath
Object localization is an important computer vision problem with a variety of applications.
Ranked #4 on Weakly Supervised Object Detection on MS COCO
no code implementations • 10 Nov 2015 • Hyungtae Lee, Heesung Kwon, Ryan M. Robinson, William D. Nothwang, Amar M. Marathe
A novel approach for the fusion of heterogeneous object detection methods is proposed.
no code implementations • 8 Jun 2015 • Zhimin Peng, Prudhvi Gurram, Heesung Kwon, Wotao Yin
In this paper, a novel framework of sparse kernel learning for Support Vector Data Description (SVDD) based anomaly detection is presented.