Rebekah Bogdanoff PRO
DualityAI-RebekahBogdanoff
AI & ML interests
None yet
Recent Activity
liked a dataset 14 days ago
Simuletic/Airport_Drone_Threat_And_Safety_Dataset updated a dataset 19 days ago
DualityAI-RebekahBogdanoff/DroneSnowAndRain liked a dataset 22 days ago
DualityAI-RebekahBogdanoff/DroneSnowAndRainOrganizations
replied to their post 22 days ago
I made at least one of the datasets here: https://huggingface.co/datasets/DualityAI-RebekahBogdanoff/DroneSnowAndRain
I would love to find some real world images that have snow/rain as occlusions though, in order to validate. Unfortunately, this truly seems like a gap
replied to their post 22 days ago
The synthetic dataset is here, if anyone is interested: https://huggingface.co/datasets/DualityAI-RebekahBogdanoff/DroneSnowAndRain
posted an update 23 days ago
Post
75
It seems like there's gaps in available drone detection data; I'm curious what you guys think are the biggest gaps? Use emojis to vote or comment a new gap:
๐ - Labels for specific drone types (different DJI drones)
๐คฏ- Rain/Snow
๐ฅ - Fog
๐ - Distractors like birds and airplanes
๐ - Long-range images/small pixels on target
What else?
๐ - Labels for specific drone types (different DJI drones)
๐คฏ- Rain/Snow
๐ฅ - Fog
๐ - Distractors like birds and airplanes
๐ - Long-range images/small pixels on target
What else?
upvoted an article 27 days ago
Article
Detecting Beyond Sight: Building AI-Enabled SAR Intelligence with Synthetic Data
DualityAI-RebekahBogdanoff
โข โข 5Post
99
Case Study: Manufacturing QA: When Your Production Run Is Shorter Than Your Training Data Collection
Modern manufacturing moves fast. Can AI-based QA system be ready before every new run โ even if the line reconfigures every day?
Read the findings here: https://huggingface.co/blog/DualityAI-RebekahBogdanoff/manufacturing-qa
Modern manufacturing moves fast. Can AI-based QA system be ready before every new run โ even if the line reconfigures every day?
Read the findings here: https://huggingface.co/blog/DualityAI-RebekahBogdanoff/manufacturing-qa
posted an update 27 days ago
Post
99
Case Study: Manufacturing QA: When Your Production Run Is Shorter Than Your Training Data Collection
Modern manufacturing moves fast. Can AI-based QA system be ready before every new run โ even if the line reconfigures every day?
Read the findings here: https://huggingface.co/blog/DualityAI-RebekahBogdanoff/manufacturing-qa
Modern manufacturing moves fast. Can AI-based QA system be ready before every new run โ even if the line reconfigures every day?
Read the findings here: https://huggingface.co/blog/DualityAI-RebekahBogdanoff/manufacturing-qa
upvoted an article 27 days ago
Article
Manufacturing QA: When Your Production Run Is Shorter Than Your Training Data Collection
DualityAI-RebekahBogdanoff
โข โข 1published an article 27 days ago
Article
Manufacturing QA: When Your Production Run Is Shorter Than Your Training Data Collection
DualityAI-RebekahBogdanoff
โข โข 1Post
142
๐ The long awaited Ground Cover Segmentation Challenge, by Duality AI + LunateAI has finally launched, along with a new feature - Dualityโs new Vibe Sim Tool! (See the comments for more info)
Vibe Sim is a new agentic simulation co-pilot that lets AI teams build scenarios, generate synthetic data, and iterate on vision models through a natural-language interface, entirely in the browser.
And I'm so excited to say that itโs now available for free to EDU users!
๐ Use Vibe Sim to Generate segmentation data to train a model to segment aerial images into the following classes:
๐ณ Trees
๐ง Water
๐ซ Buildings
๐ผ๏ธ Other
Ground cover segmentation is an important task for geospatial analysis of human environmental impact, deforestation, and overall geospatial changes over time, but anyone who has ever had to segment an image for model training understands how much time and effort is saved by the automatic labeling ๐ฎโ๐จ
Synthetic data from simulation allows users to train a model that works on real world images, using automatically annotated data that bridges the Sim2Real gap.
Join this challenge to learn how to do just that!
Utilize Dualityโs full twin library, including geospatial environment twins based on real-world locations. Customize simulations using a natural-language interface and quickly iterate through the training/testing process to improve your score!
Duality is excited to once again be partnering with LunateAI, a high-end advisory business founded by the award-winning, industry-recognized global leader Dr. Nadine Alameh to usher in a new era of geospatial impact in conjunction with advances in computing and AI.
LunateAI helps government and industry leaders ๐ค rethink, ๐กredesign, and ๐ execute transformative geospatial strategies using AI, cloud, and Lunateโs unparalleled global expertise.
Vibe Sim is a new agentic simulation co-pilot that lets AI teams build scenarios, generate synthetic data, and iterate on vision models through a natural-language interface, entirely in the browser.
And I'm so excited to say that itโs now available for free to EDU users!
๐ Use Vibe Sim to Generate segmentation data to train a model to segment aerial images into the following classes:
๐ณ Trees
๐ง Water
๐ซ Buildings
๐ผ๏ธ Other
Ground cover segmentation is an important task for geospatial analysis of human environmental impact, deforestation, and overall geospatial changes over time, but anyone who has ever had to segment an image for model training understands how much time and effort is saved by the automatic labeling ๐ฎโ๐จ
Synthetic data from simulation allows users to train a model that works on real world images, using automatically annotated data that bridges the Sim2Real gap.
Join this challenge to learn how to do just that!
Utilize Dualityโs full twin library, including geospatial environment twins based on real-world locations. Customize simulations using a natural-language interface and quickly iterate through the training/testing process to improve your score!
Duality is excited to once again be partnering with LunateAI, a high-end advisory business founded by the award-winning, industry-recognized global leader Dr. Nadine Alameh to usher in a new era of geospatial impact in conjunction with advances in computing and AI.
LunateAI helps government and industry leaders ๐ค rethink, ๐กredesign, and ๐ execute transformative geospatial strategies using AI, cloud, and Lunateโs unparalleled global expertise.
replied to their post 5 months ago
replied to their post 5 months ago
posted an update 5 months ago
Post
142
๐ The long awaited Ground Cover Segmentation Challenge, by Duality AI + LunateAI has finally launched, along with a new feature - Dualityโs new Vibe Sim Tool! (See the comments for more info)
Vibe Sim is a new agentic simulation co-pilot that lets AI teams build scenarios, generate synthetic data, and iterate on vision models through a natural-language interface, entirely in the browser.
And I'm so excited to say that itโs now available for free to EDU users!
๐ Use Vibe Sim to Generate segmentation data to train a model to segment aerial images into the following classes:
๐ณ Trees
๐ง Water
๐ซ Buildings
๐ผ๏ธ Other
Ground cover segmentation is an important task for geospatial analysis of human environmental impact, deforestation, and overall geospatial changes over time, but anyone who has ever had to segment an image for model training understands how much time and effort is saved by the automatic labeling ๐ฎโ๐จ
Synthetic data from simulation allows users to train a model that works on real world images, using automatically annotated data that bridges the Sim2Real gap.
Join this challenge to learn how to do just that!
Utilize Dualityโs full twin library, including geospatial environment twins based on real-world locations. Customize simulations using a natural-language interface and quickly iterate through the training/testing process to improve your score!
Duality is excited to once again be partnering with LunateAI, a high-end advisory business founded by the award-winning, industry-recognized global leader Dr. Nadine Alameh to usher in a new era of geospatial impact in conjunction with advances in computing and AI.
LunateAI helps government and industry leaders ๐ค rethink, ๐กredesign, and ๐ execute transformative geospatial strategies using AI, cloud, and Lunateโs unparalleled global expertise.
Vibe Sim is a new agentic simulation co-pilot that lets AI teams build scenarios, generate synthetic data, and iterate on vision models through a natural-language interface, entirely in the browser.
And I'm so excited to say that itโs now available for free to EDU users!
๐ Use Vibe Sim to Generate segmentation data to train a model to segment aerial images into the following classes:
๐ณ Trees
๐ง Water
๐ซ Buildings
๐ผ๏ธ Other
Ground cover segmentation is an important task for geospatial analysis of human environmental impact, deforestation, and overall geospatial changes over time, but anyone who has ever had to segment an image for model training understands how much time and effort is saved by the automatic labeling ๐ฎโ๐จ
Synthetic data from simulation allows users to train a model that works on real world images, using automatically annotated data that bridges the Sim2Real gap.
Join this challenge to learn how to do just that!
Utilize Dualityโs full twin library, including geospatial environment twins based on real-world locations. Customize simulations using a natural-language interface and quickly iterate through the training/testing process to improve your score!
Duality is excited to once again be partnering with LunateAI, a high-end advisory business founded by the award-winning, industry-recognized global leader Dr. Nadine Alameh to usher in a new era of geospatial impact in conjunction with advances in computing and AI.
LunateAI helps government and industry leaders ๐ค rethink, ๐กredesign, and ๐ execute transformative geospatial strategies using AI, cloud, and Lunateโs unparalleled global expertise.