1,000+ Opportunities
Find the right grant
Search federal, foundation, and corporate grants with AI — or browse by agency, topic, and state.
Submission period runs April 3 – July 31, 2026. Development dataset released April 4, challenge dataset released June 1. Winner announcement August 9-14 at IGARSS 2026.
2026 WRIVA Cross-View Geo-Localization (CVGL) Challenge is a grant from the Intelligence Advanced Research Projects Activity (IARPA) that funds teams competing to advance the state of the art in associating ground-level images with satellite and aerial overhead observations for fine-grained geolocation.
The challenge addresses near-orthogonal viewpoint differences, dramatic scale disparities, and occlusion effects that limit ground-to-satellite matching accuracy. Applications span geo-referencing of unlocalized photos, satellite data attribution, 3D scene reconstruction, digital twins, and environmental monitoring.
The competition, hosted on IEEE DataPort, invites multidisciplinary teams from academia and industry with expertise in deep learning, cross-view retrieval, and multi-modal representation learning. Prize details and registration information are available on the IEEE DataPort challenge page.
Get alerted about grants like this
Get emailed when new opportunities from “Intelligence Advanced Research Projects Activity (IARPA)” or related funders appear. Free, weekly, unsubscribe anytime.
Or search similar grants →Extracted from the official opportunity page/RFP to help you evaluate fit faster.
WRIVA CVGL Challenge 2026 | IEEE DataPort Geoscience and Remote Sensing Ground-level images provide unique contextual and semantic information that complements satellite and aerial observation. Establishing reliable connections between these modalities - addressing challenges due to near-orthogonal viewpoint differences, dramatic scale and resolution disparities, and occlusion effects - unlocks a broad range of emerging applications.
These include fine-grained geo-referencing of unlocalized photos, satellite data attribution, 3D scene reconstruction across varying altitudes, multi-view data fusion for digital twins, and cross-source change detection for environmental and infrastructure monitoring.
Recent advances in deep learning, cross-view retrieval, and multi-modal representation learning have markedly improved our ability to associate ground images with overhead observations; however, the task remains extremely challenging: differences in geometry, radiometry, and content visibility persist, and globally accurate ground-to-satellite localization is still limited to tens of meters without additional priors.
The next frontier lies in local-context geo-localization, where approximate camera positions (e.g., within a few city blocks) enable meter-level alignment and open opportunities for downstream mapping, navigation, and situational awareness.
This cross-view geo-localization (CVGL) competition is sponsored by the Intelligence Advanced Research Projects Activity (IARPA) Walk-through Rendering from Images of Varying Altitude (WRIVA) program. It seeks to promote innovation and reproducibility for local-context ground-to-satellite image localization.
Participants will receive sets of one or more satellite images and one or more ground-level images with approximate locations (e.g., within a few hundred meters) and will be evaluated on predicted camera geolocation accuracy in meters.
Winning teams will be acknowledged during a session at IGARSS 2026 and may be invited to contribute to an article on the competition to be published in a special issue of Photogrammetric Engineering & Remote Sensing (PE&RS).
Development dataset release on DataPort: 4 April 2026 Baseline and metrics code release on GitHub: 13 April 2026 Development leaderboard posted on CodaBench: NLT 1 May 2026 Challenge dataset release on DataPort: 1 June 2026 Challenge submission period: June 2026 – July 2026 Selection of winning participants: August 2026 Announcement of winners at IGARSS 2026: 9-14 August 2026 Given multiple groups of sequential ground-level images and one or more satellite images covering the full site, locate each camera accurately.
Ground-level images are intentionally placed such that camera calibration of all images in a scene without cross-view matching using the satellite images is difficult. Multiple datasets of varying difficulty will be provided.
Unposed ground-level images Multiple orthorectified Maxar satellite images with GeoTIFF metadata JSON text file including camera locations for each input ground-level image Evaluation metric for ground-level camera geo-localization Horizontal position error (meters), ninetieth percentile for each dataset, averaged over all datasets and reported with two significant digits The leaderboard metric may be refined throughout the development phase in advance of launching the test phase Contest Data and Public Datasets for Algorithm Development WRIVA-CVGL-DEV: Images collected for the IARPA WRIVA program have been publicly released and made available for use in this public challenge and more broadly to encourage research in cross-view camera calibration, geo-localization, and view synthesis methods for real-world environments and heterogeneous cameras.
Ground-level images were collected using a variety of cameras equipped with RTK-capable GPS receivers. Cameras were calibrated using structure from motion constrained by RTK-corrected GPS coordinates, with positioning accuracies measured in centimeters. Orthorectified satellite images are (c) 2026 Maxar.
A sample of WRIVA-CVGL datasets with reference metadata is provided to validate algorithms during the development phase. Challenge datasets from WRIVA-CVGL will be provided without metadata. VisymScenes-CVGL: Public datasets suitable for metric CVGL include DReSS, VIGOR, Ford-CVL, and KITTI-CVL, among others.
These datasets provide satellite images from Google or ESRI and ground level panoramas or individual image frames, each with reasonably accurate camera location metadata. Ground-level photos tend to be taken from street view, which may not be representative of use with hand-held photography.
Satellite images tend to be of high visual quality taken in temperate seasons, which may not be representative of timely satellite images available in practice. To provide additional, more representative data for our task, we provide multiple orthorectified satellite images from (c) 2026 Maxar to augment the large-scale public VisymScenes hand-held photography dataset (https://huggingface. co/datasets/doppelgangers25/VisymScenes).
Camera location accuracies as reported from the mobile phones used to collect the images are indicated in the VisymScenes image metadata. Participants in the challenge may use any publicly available data to develop their algorithms. Caveat: Satellite image metadata for WRIVA-CVGL-DEV and VisymScenes-CVGL have not been bundle adjusted or refined with ground control.
Absolute and relative accuracy is expected to be within a few meters. This will limit accuracy of metrics calculations for the development datasets to no better than a few meters. By participating in this challenge, you consent to publicly share your submissions.
No financial prizes will be awarded. Organizers will review submissions and invite top-scoring participants to submit a brief writeup of their solutions, documenting their approaches and observations. Organizers will then select from among the best scores and writeups and acknowledge winners in a session at IGARSS 2026.
All top-scoring participants on the leaderboard will be acknowledged. Winning teams may also be invited to contribute to an article on the competition to be published in a special issue of Photogrammetric Engineering & Remote Sensing (PE&RS). Contest data is available for download from this DataPort repository.
Development data is currently provided. Challenge data will be provided at the beginning of the challenge phase. Submissions will be managed with a CodaBench leaderboard.
A link will be added soon. An example baseline solution and metrics code will be maintained on GitHub (https://github. com/pubgeo/wriva-cvgl-baseline) and improved throughout the development phase.
While we do not expect the baseline to be competitive, it will provide an example of using the provided development data inputs and producing outputs in the format expected for metric evaluation. Additional details about output file expectations will be provided upon leaderboard launch. A copy of the baseline code, data split pickle files, and model weights dated 4/13 is provided for download here since the non-code files are large.
For the most recent code and documentation, please see the GitHub repo. This work was supported by the Intelligence Advanced Research Projects Activity (IARPA) Walk-through Rendering from Images of Varying Altitude (WRIVA) program.
Disclaimer: The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of IARPA or the U.S. Government. Can you release checksums (e.g. md5 or sha256) for the dataset zip files? In reply to Can you release checksums (e… by Michael Chaberski Sure thing.
I just uploaded a text file with the MD5s for the current zips. When we add new zips, we'll plan to update that file with any new checksums. In reply to Sure thing.
I just uploaded… by Myron Brown VisymScenes-CVGL-001-020. zip (Size: 11. 14 GB) VisymScenes-CVGL-021-040.
zip (Size: 18 GB) VisymScenes-CVGL-041-060. zip (Size: 13. 68 GB) VisymScenes-CVGL-061-080.
zip (Size: 16. 58 GB) VisymScenes-CVGL-081-100. zip (Size: 16.
84 GB) VisymScenes-CVGL-101-120. zip (Size: 12 GB) VisymScenes-CVGL-121-140. zip (Size: 12.
35 GB) VisymScenes-CVGL-141-149. zip (Size: 3. 67 GB) WRIVA-CVGL-DEV.
zip (Size: 7. 05 GB) 260413-ZIP-MD5s. txt (Size: 731 bytes) 260413-wriva-cvgl-initial-baseline.
zip (Size: 619. 82 MB) You must be an approved participant in this data competition to access dataset files. To request access you must first " class="btn btn-primary">Login 260413-ZIP-MD5s_0.
txt (731 bytes) 260413-wriva-cvgl-initial-baseline_0. zip (619. 82 MB) VisymScenes-CVGL-001-020.
zip (11. 14 GB) VisymScenes-CVGL-021-040. zip (18 GB) VisymScenes-CVGL-041-060.
zip (13. 68 GB) VisymScenes-CVGL-061-080. zip (16.
58 GB) VisymScenes-CVGL-081-100. zip (16. 84 GB) VisymScenes-CVGL-101-120.
zip (12 GB) VisymScenes-CVGL-121-140. zip (12. 35 GB) VisymScenes-CVGL-141-149.
zip (3. 67 GB) WRIVA-CVGL-DEV. zip (7.
05 GB) You must be an approved participant in this data competition to access dataset files.
To request access you must first " class="btn btn-primary">Login Send Author a Private Message Report a problem with this Dataset How to Upload Dataset Files Directly to AWS Upload Your Files directly to the IEEE DataPort S3 Bucket You will need the following information to complete your upload: Your AWS Access Key and Secret key, which can be found on your IEEE DataPort User Profile .
Homer Li, Jessica Ye, Neil Joshi, Joshua Carney, Rongjun Qin, Myron Brown, "WRIVA CVGL Challenge 2026", IEEE Dataport, April 3, 2026, doi:10. 21227/y7nm-c095 {https://dx. doi.
org/10. 21227/y7nm-c095}, {Homer Li and Jessica Ye and Neil Joshi and Joshua Carney and Rongjun Qin and Myron Brown}, publisher = {IEEE Dataport}, {WRIVA CVGL Challenge 2026}, WRIVA CVGL Challenge 2026 Homer Li; Jessica Ye; Neil Joshi; Joshua Carney; Rongjun Qin; Myron Brown Homer Li, Jessica Ye, Neil Joshi, Joshua Carney, Rongjun Qin, Myron Brown WRIVA CVGL Challenge 2026. IEEE Dataport.
https://dx. doi. org/10.
21227/y7nm-c095 Homer Li, Jessica Ye, Neil Joshi, Joshua Carney, Rongjun Qin, Myron Brown WRIVA CVGL Challenge 2026. Available at: https://dx. doi.
org/10. 21227/y7nm-c095 Homer Li, Jessica Ye, Neil Joshi, Joshua Carney, Rongjun Qin, Myron Brown. (2026).
"WRIVA CVGL Challenge 2026." Web, Homer Li, Jessica Ye, Neil Joshi, Joshua Carney, Rongjun Qin, Myron Brown. WRIVA CVGL Challenge 2026 [Internet].
IEEE Dataport; https://dx. doi. org/10.
21227/y7nm-c095 Homer Li, Jessica Ye, Neil Joshi, Joshua Carney, Rongjun Qin, Myron Brown, "WRIVA CVGL Challenge 2026,", Embed this dataset on another website Copy and paste the HTML code below to embed your dataset: Share a link to this dataset https://ieee-dataport. org//competitions/wriva-cvgl-challenge-2026 https://dx. doi.
org/10. 21227/y7nm-c095 Urban Semantic 3D Dataset
According to the current listing, eligibility includes: Open to researchers who consent to publicly share submissions; dataset access requires approval. Expertise in computer vision and geo-localization expected. Confirm the full requirements in the official notice before applying.
Applications for 2026 WRIVA Cross-View Geo-Localization (CVGL) Challenge are due July 31, 2026. Build your timeline backwards from this date to cover registrations, approvals, and final submission checks.
2026 WRIVA Cross-View Geo-Localization (CVGL) Challenge is funded by Intelligence Advanced Research Projects Activity (IARPA). Verify program details on the funder's official page before applying.
This listing is flagged as international in scope. Check the official notice for country-specific restrictions before applying.
Applications go through the funder's official portal — the Apply Now link on this page goes there directly.
Commercial Observation for Spatio-Temporal Monitoring for Indications of Change (COSMIC) is sponsored by Intelligence Advanced Research Projects Activity (IARPA). The COSMIC program seeks to develop a methodology to leverage commercial remote sensing technologies and open-source geolocation information to generate pseudo-persistent data (PPD). COSMIC aims to create a methodology to rapidly integrate multiple sources of information into a unified framework to allow for incorporation into a layered GEOINT model or PPD. This framework should also allow the prediction and visualization of resolution and bands of data not provided for PPD construction through predictive processes such as image diffusion or other generative AI approaches.
IARPA COSMIC Commercial Observation AI for Spatio-temporal Geospatial Intelligence is sponsored by Intelligence Advanced Research Projects Activity (IARPA). IARPA's Commercial Observation for Spatio-temporal Monitoring for Indications of Change (COSMIC) program aims to integrate commercial remote sensing data and open-source geolocation information into dynamic geospatial models and develop an agentic AI analytic system capable of a…
IARPA's Commercial Observation for Spatio-temporal Monitoring for Indications of Change (COSMIC) program aims to integrate commercial remote sensing data and open-source geolocation information into dynamic geospatial models and develop an agentic AI analytic system capable of answering intelligence questions. The program is part of IARPA's Emerging Technology Accelerator (ETA) framework released in April 2026, supporting high-risk high-payoff R&D for the Intelligence Community. COSMIC seeks to create AI systems that generate pseudo-persistent geospatial intelligence from non-persistent commercial satellite and sensor data.
DoD Multidisciplinary Research Program of the University Research Initiative (MURI) is sponsored by Department of Defense (DoD) - Office of Naval Research (ONR). The Multidisciplinary Research Program of the University Research Initiative (MURI), administered by the Department of Defense Office of Naval Research, supports basic research in science and engineering at U. S.
SBIR SF254-D1206: Knowledge-Guided Test and Evaluation Frameworks for proliferated Low Earth Orbit Constellations is sponsored by U.S. Air Force. DOD SBIR topic SF254-D1206: Knowledge-Guided Test and Evaluation Frameworks for proliferated Low Earth Orbit Constellations. Component: U.S. Air Force. Command: SDA. Solicitation: DoD SBIR 2025.4. Phase(s): D2PII, II, SPII. Status: Pre-Release. Open date: 3/4/2026.
The FY2026 federal funding map has tilted hard toward AI, critical minerals, energy, advanced manufacturing, and workforce development — while a new layer of political review asks whether each award advances administration priorities. Here is a strategic map of where the money is moving, and how to position a proposal for the new alignment screen without distorting the work.
Read articleTitle III of the Defense Production Act lets the Pentagon hand non-dilutive capital to companies that expand domestic production of defense-critical materials and manufacturing — through a white-paper-first pathway run by the Air Force Research Laboratory (FA8650-19-S-5010) rather than a conventional grant competition. But the core DPA authorities sunset September 30, 2026 absent reauthorization, and the standing white-paper window has moved in and out of suspension. Here is how the Title III mechanism actually works, why it rewards companies that lead with a supply-chain vulnerability, and how to position before the authority cliff.
Read articleNSF's NAIRR Operations Center solicitation (NSF 25-546) will make a single award of up to $35 million over five years to the organization that will run the National AI Research Resource — the shared computing, data, and model infrastructure that has already connected over 400 U.S. research teams. This is a rare winner-take-all federal competition where the prize is not funding for your own research but the mandate to operate national infrastructure. Here is what the NAIRR is, why its transition from pilot to permanent program matters, and what kind of organization can credibly win it.
Read article