Search Results for author: Nathan Jacobs

Found 43 papers, 11 papers with code

Learning Tri-modal Embeddings for Zero-Shot Soundscape Mapping

no code implementations19 Sep 2023 Subash Khanal, Srikumar Sastry, Aayush Dhakal, Nathan Jacobs

We focus on the task of soundscape mapping, which involves predicting the most probable sounds that could be perceived at a particular geographic location.

 Ranked #1 on Cross-Modal Retrieval on SoundingEarth (using extra training data)

Cross-Modal Retrieval

StereoFlowGAN: Co-training for Stereo and Flow with Unsupervised Domain Adaptation

no code implementations4 Sep 2023 Zhexiao Xiong, Feng Qiao, Yu Zhang, Nathan Jacobs

We introduce a novel training strategy for stereo matching and optical flow estimation that utilizes image-to-image translation between synthetic and real image domains.

Image-to-Image Translation Optical Flow Estimation +3

Fine-Grained Property Value Assessment using Probabilistic Disaggregation

no code implementations31 May 2023 Cohen Archbold, Benjamin Brodie, Aram Ansary Ogholbake, Nathan Jacobs

The monetary value of a given piece of real estate, a parcel, is often readily available from a geographic information system.

A Visual Active Search Framework for Geospatial Exploration

1 code implementation28 Nov 2022 Anindya Sarkar, Michael Lanier, Scott Alfeld, Jiarui Feng, Roman Garnett, Nathan Jacobs, Yevgeniy Vorobeychik

We model this class of problems in a visual active search (VAS) framework, which takes as input an image of a broad area, and aims to identify as many examples of a target object as possible.

Domain Adaptation

Geo-Information Harvesting from Social Media Data

no code implementations1 Nov 2022 Xiao Xiang Zhu, Yuanyuan Wang, Mrinalini Kochupillai, Martin Werner, Matthias Häberle, Eike Jens Hoffmann, Hannes Taubenböck, Devis Tuia, Alex Levering, Nathan Jacobs, Anna Kruspe, Karam Abdulahhad

In this article, we address key aspects in the field, including data availability, analysis-ready data preparation and data management, geo-information extraction from social media text messages and images, and the fusion of social media and remote sensing data.


Causality for Inherently Explainable Transformers: CAT-XPLAIN

1 code implementation29 Jun 2022 Subash Khanal, Benjamin Brodie, Xin Xing, Ai-Ling Lin, Nathan Jacobs

There have been several post-hoc explanation approaches developed to explain pre-trained black-box neural networks.

Binary Classification

Revisiting Near/Remote Sensing with Geospatial Attention

no code implementations CVPR 2022 Scott Workman, M. Usman Rafique, Hunter Blanton, Nathan Jacobs

We introduce a novel architecture for near/remote sensing that is based on geospatial attention and demonstrate its use for five segmentation tasks.

Image Segmentation Semantic Segmentation

Dynamic Feature Alignment for Semi-supervised Domain Adaptation

no code implementations18 Oct 2021 Yu Zhang, Gongbo Liang, Nathan Jacobs

Most research on domain adaptation has focused on the purely unsupervised setting, where no labeled examples in the target domain are available.

Domain Adaptation Semi-supervised Domain Adaptation

Intensity Harmonization for Airborne LiDAR

no code implementations4 May 2021 David Jones, Nathan Jacobs

We show that our method performs as well as the best baseline in areas with similar intensity distributions, and outperforms all baselines in areas with different intensity distributions.

Towards a Collective Agenda on AI for Earth Science Data Analysis

1 code implementation11 Apr 2021 Devis Tuia, Ribana Roscher, Jan Dirk Wegner, Nathan Jacobs, Xiao Xiang Zhu, Gustau Camps-Valls

In the last years we have witnessed the fields of geosciences and remote sensing and artificial intelligence to become closer.

Learning a Dynamic Map of Visual Appearance

no code implementations CVPR 2020 Tawfiq Salem, Scott Workman, Nathan Jacobs

The appearance of the world varies dramatically not only from place to place but also from hour to hour and month to month.

A Structure-Aware Method for Direct Pose Estimation

no code implementations22 Dec 2020 Hunter Blanton, Scott Workman, Nathan Jacobs

Direct methods, such as PoseNet, regress pose from the image as a fixed function, for example using a feed-forward convolutional network.

Image Retrieval Pose Estimation +2

Dynamic Traffic Modeling From Overhead Imagery

no code implementations CVPR 2020 Scott Workman, Nathan Jacobs

Our goal is to use overhead imagery to understand patterns in traffic flow, for instance answering questions such as how fast could you traverse Times Square at 3am on a Sunday.

Optical Wavelength Guided Self-Supervised Feature Learning For Galaxy Cluster Richness Estimate

no code implementations4 Dec 2020 Gongbo Liang, Yuanyuan Su, Sheng-Chieh Lin, Yu Zhang, Yuanyuan Zhang, Nathan Jacobs

We believe the proposed method will benefit astronomy and cosmology, where a large number of unlabeled multi-band images are available, but acquiring image labels is costly.


Contrastive Cross-Modal Pre-Training: A General Strategy for Small Sample Medical Imaging

no code implementations6 Oct 2020 Gongbo Liang, Connor Greenwell, Yu Zhang, Xiaoqin Wang, Ramakanth Kavuluru, Nathan Jacobs

A key challenge in training neural networks for a given medical imaging task is often the difficulty of obtaining a sufficient number of manually labeled examples.

Image Classification Image-text matching +2

Improved Trainable Calibration Method for Neural Networks on Medical Imaging Classification

no code implementations9 Sep 2020 Gongbo Liang, Yu Zhang, Xiaoqin Wang, Nathan Jacobs

Recent works have shown that deep neural networks can achieve super-human performance in a wide range of image classification tasks in the medical imaging domain.

Classification Decision Making +2

Single Image Cloud Detection via Multi-Image Fusion

no code implementations29 Jul 2020 Scott Workman, M. Usman Rafique, Hunter Blanton, Connor Greenwell, Nathan Jacobs

A primary challenge in developing algorithms for identifying such artifacts is the cost of collecting annotated training data.

Cloud Detection object-detection +2

Estimating Displaced Populations from Overhead

1 code implementation25 Jun 2020 Armin Hadzic, Gordon Christie, Jeffrey Freeman, Amber Dismer, Stevan Bullard, Ashley Greiner, Nathan Jacobs, Ryan Mukherjee

We introduce a deep learning approach to perform fine-grained population estimation for displacement camps using high-resolution overhead imagery.


Defense-PointNet: Protecting PointNet Against Adversarial Attacks

no code implementations27 Feb 2020 Yu Zhang, Gongbo Liang, Tawfiq Salem, Nathan Jacobs

Despite remarkable performance across a broad range of tasks, neural networks have been shown to be vulnerable to adversarial attacks.

Adversarial Robustness

2D Convolutional Neural Networks for 3D Digital Breast Tomosynthesis Classification

no code implementations27 Feb 2020 Yu Zhang, Xiaoqin Wang, Hunter Blanton, Gongbo Liang, Xin Xing, Nathan Jacobs

Automated methods for breast cancer detection have focused on 2D mammography and have largely ignored 3D digital breast tomosynthesis (DBT), which is frequently used in clinical practice.

Breast Cancer Detection Classification +1

Learning Geo-Temporal Image Features

no code implementations16 Sep 2019 Menghua Zhai, Tawfiq Salem, Connor Greenwell, Scott Workman, Robert Pless, Nathan Jacobs

We propose to implicitly learn to extract geo-temporal image features, which are mid-level features related to when and where an image was captured, by explicitly optimizing for a set of location and time estimation tasks.

Learning to Map Nearly Anything

no code implementations16 Sep 2019 Tawfiq Salem, Connor Greenwell, Hunter Blanton, Nathan Jacobs

Looking at the world from above, it is possible to estimate many properties of a given location, including the type of land cover and the expected land use.

Remote Estimation of Free-Flow Speeds

no code implementations24 Jun 2019 Weilian Song, Tawfiq Salem, Hunter Blanton, Nathan Jacobs

We propose an automated method to estimate a road segment's free-flow speed from overhead imagery and road metadata.

FARSA: Fully Automated Roadway Safety Assessment

1 code implementation17 Jan 2019 Weilian Song, Scott Workman, Armin Hadzic, Xu Zhang, Eric Green, Mei Chen, Reginald Souleyrette, Nathan Jacobs

An emerging approach for conducting such assessments in the United States is through the US Road Assessment Program (usRAP), which rates roads from highest risk (1 star) to lowest (5 stars).

A Weakly Supervised Approach for Estimating Spatial Density Functions from High-Resolution Satellite Imagery

1 code implementation22 Oct 2018 Nathan Jacobs, Adam Kraft, Muhammad Usman Rafique, Ranti Dev Sharma

We propose a neural network component, the regional aggregation layer, that makes it possible to train a pixel-level density estimator using only coarse-grained density aggregates, which reflect the number of objects in an image region.

What Goes Where: Predicting Object Distributions from Above

no code implementations2 Aug 2018 Connor Greenwell, Scott Workman, Nathan Jacobs

In this work, we propose a cross-view learning approach, in which images captured from a ground-level view are used as weakly supervised annotations for interpreting overhead imagery.

Learning to Look around Objects for Top-View Representations of Outdoor Scenes

no code implementations ECCV 2018 Samuel Schulter, Menghua Zhai, Nathan Jacobs, Manmohan Chandraker

Given a single RGB image of a complex outdoor road scene in the perspective view, we address the novel problem of estimating an occlusion-reasoned semantic scene layout in the top-view.

Semantic Segmentation

A Unified Model for Near and Remote Sensing

no code implementations ICCV 2017 Scott Workman, Menghua Zhai, David J. Crandall, Nathan Jacobs

To evaluate our approach, we created a large dataset of overhead and ground-level images from a major urban area with three sets of labels: land use, building function, and building age.

Density Estimation

Revisiting IM2GPS in the Deep Learning Era

no code implementations ICCV 2017 Nam Vo, Nathan Jacobs, James Hays

The recent state-of-the-art approach to this problem is a deep image classification approach in which the world is spatially divided into cells and a deep network is trained to predict the correct cell for a given image.

Ranked #4 on Photo geolocation estimation on Im2GPS (using extra training data)

Classification Density Estimation +3

Understanding and Mapping Natural Beauty

no code implementations ICCV 2017 Scott Workman, Richard Souvenir, Nathan Jacobs

While natural beauty is often considered a subjective property of images, in this paper, we take an objective approach and provide methods for quantifying and predicting the scenicness of an image.

Detecting Vanishing Points using Global Image Context in a Non-Manhattan World

1 code implementation CVPR 2016 Menghua Zhai, Scott Workman, Nathan Jacobs

Our method reverses this process: we propose a set of horizon line candidates and score each based on the vanishing points it contains.

Horizon Line Estimation

Horizon Lines in the Wild

1 code implementation7 Apr 2016 Scott Workman, Menghua Zhai, Nathan Jacobs

The horizon line is an important contextual attribute for a wide variety of image understanding tasks.

Horizon Line Estimation

Building Dynamic Cloud Maps From the Ground Up

no code implementations ICCV 2015 Calvin Murdock, Nathan Jacobs, Robert Pless

Satellite imagery of cloud cover is extremely important for understanding and predicting weather.

Wide-Area Image Geolocalization with Aerial Reference Imagery

no code implementations ICCV 2015 Scott Workman, Richard Souvenir, Nathan Jacobs

We propose to use deep convolutional neural networks to address the problem of cross-view image geolocalization, in which the geolocation of a ground-level query image is estimated by matching to georeferenced aerial images.

Cloud Motion as a Calibration Cue

no code implementations CVPR 2013 Nathan Jacobs, Mohammad T. Islam, Scott Workman

We propose cloud motion as a natural scene cue that enables geometric calibration of static outdoor cameras.

Shadow Estimation Method for "The Episolar Constraint: Monocular Shape from Shadow Correspondence"

no code implementations15 Apr 2013 Austin Abrams, Chris Hawley, Kylia Miskell, Adina Stoica, Nathan Jacobs, Robert Pless

We show these approaches only work with very careful tuning of parameters, and do not work well for long-term time-lapse sequences taken over the span of many months.

Cannot find the paper you are looking for? You can Submit a new open access paper.