You can also find my articles on my Google Scholar profile.
Published in arxiv preprint, 2020
We jointly predict depth and scene coordinates directly from a single RGB image and use the resulting corresponding point clouds to regress camera pose.
Recommended citation: Hunter Blanton, Scott Workman, Nathan Jacobs. "A Structure-Aware Method for Direct Pose Estimation." arxiv. 2020. https://arxiv.org/abs/2012.12360
Published in IEEE International Geoscience and Remote Sensing Symposium, 2020
We use a a mutli-image fusion training approach and adapt it to single image cloud detection.
Recommended citation: Scott Workman, M. Usman Rafique, Hunter Blanton, Connor Greenwell, Nathan Jacobs. "Single Image Cloud Detection via Multi-Image Fusion", IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2020.
Published in IEEE International Geoscience and Remote Sensing Symposium, 2020
We use lidar points and scan ray data to learn a lidar surface model that can be used to estimate novel points from single scans.
Recommended citation: Hunter Blanton, Sean Grate, Nathan Jacobs. "Surface Modeling for Airborne Lidar", IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2020.
Published in British Machine Vision Conference, 2020
We use spatially paired streetview panoramas to train a network single image view synthesis.
Recommended citation: Muhammad Usman Rafique, Hunter Blanton, Noah Snavely, Nathan Jacobs. "Generative Appearance Flow: A Hybrid Approach for Outdoor View Synthesis." British Machine Vision Conference (BMVC). 2020.
Published in ECCV 2020 Workshop on BioImage Computing, 2020
Construct dynamic images from 3D MRIs of the brain to use with 2D CNNs for Alshiemer’s classification.
Recommended citation: Xin Xing, Gongbo Liang, Hunter Blanton, M. Usman Rafique, Chris Wang, Ai-Ling Lin, Nathan Jacobs. "Dynamic Image for 3D MRI image Alzheimer’s Disease classification." ECCV Workshop on BioImage Computing. 2020. https://openreview.net/pdf?id=HdNVXBdk05
Published in CVPR Workshop: Large Scale Computer Vision for Remote Sensing Imagery (EARTHVISION), 2020
Free-flow speed estimation using combined features from point clouds and overhead imagery.
Recommended citation: Hadzic, Armin, et al. "RasterNet: Modeling Free-Flow Speed Using LiDAR and Overhead Imagery." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. 2020. http://openaccess.thecvf.com/content_CVPRW_2020/papers/w11/Hadzic_RasterNet_Modeling_Free-Flow_Speed_Using_LiDAR_and_Overhead_Imagery_CVPRW_2020_paper.pdf
Published in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2020
We extend PoseNet for localization in several scenes using a single CNN.
Recommended citation: Blanton, Hunter, et al. "Extending Absolute Pose Regression to Multiple Scenes." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. 2020. http://openaccess.thecvf.com/content_CVPRW_2020/papers/w3/Blanton_Extending_Absolute_Pose_Regression_to_Multiple_Scenes_CVPRW_2020_paper.pdf
Published in Journal of the American College of Radiology, 2020
We show that various deep learning models exhibit inconsistent performance between data sources.
Recommended citation: Wang, Xiaoqin, et al. "Inconsistent Performance of Deep Learning Models on Mammogram Classification." Journal of the American College of Radiology (2020).
Published in IEEE International Conference on Bioinformatics and Biomedicine, 2019
We use both 2D and 3D imaging modalities for breast cancer classification.
Recommended citation: Liang, Gongbo, et al. "Joint 2D-3D Breast Cancer Classification." 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). IEEE, 2019. https://arxiv.org/pdf/2002.12392.pdf
Published in IEEE International Conference on Bioinformatics and Biomedicine, 2019
We use 2D CNNs for breast cancer classification in 3D DBT images.
Recommended citation: Zhang, Yu, et al. "2D Convolutional Neural Networks for 3D Digital Breast Tomosynthesis Classification." 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). IEEE, 2019.
Published in IEEE International Geoscience and Remote Sensing Symposium, 2019
We estimate road free-flow speed directly from imagery.
Recommended citation: Song, Weilian, et al. "Remote estimation of free-flow speeds." IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2019. https://arxiv.org/pdf/1906.10104.pdf
Published in IEEE International Geoscience and Remote Sensing Symposium, 2019
We jointly learn a variety of distributions over objects and scene categories from overhead images using paired ground level image.
Recommended citation: MSalem, Tawfiq, et al. "Learning to map nearly anything." IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2019. https://arxiv.org/pdf/1909.06928.pdf
Published in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2019
Fuse multiple images without labels
Recommended citation: Muhammad Usman Rafique, Hunter Blanton, Nathan Jacobs; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2019, pp. 0-0 http://openaccess.thecvf.com/content_CVPRW_2019/html/EarthVision/Rafique_Weakly_Supervised_Fusion_of_Multiple_Overhead_Images_CVPRW_2019_paper.html