Nobody owns this listing yet

Dobb-E is live in the directory, but no verified owner controls the page.

Claim this listing
  • Pick your launch daySchedule a launch and compete for the daily #1 spot.
  • Win trophies you can embedRank top 3 on launch day and get an embeddable trophy.
  • “Featured on AI Kaptan” badgeOwner-only badge with your live upvote count.
  • Control what the page saysEdit copy, pricing, links and screenshots as the owner.

Unclaimed listings don't stay up forever. Listings left unclaimed are removed from the directory in periodic cleanups. When a listing goes, it goes everywhere — the tool page and every alternatives comparison it appears on, along with the rankings and backlinks that page had built up.

Verify with a work email on the domain (about 2 minutes), then a one-time $5 listing fee. See everything you get

Dobb-E
Robotics
Free

Dobb-E

An open-source framework to teach robots household tasks in 20 minutes using an iPhone-based demonstration tool and pretrained models.

Rating
0
Reviews
0
Upvotes
3 months ago
Listed
Robotics
Open Source
Artificial Intelligence
Machine Learning
Home Automation
Computer Vision
Research Tool
Tool information
Provider
NYU / Meta AI
Platforms
Web
Linux
iOS (for data collection)
Languages
English
Added to directory
3 months ago
Last updated
about 1 month ago
Rating
Not available
Pricing
Free

About Dobb-E

What the tool does and who it's for

Dobb-E is an open-source robotic framework designed to bridge the gap between laboratory research and real-world household automation. Developed by researchers at NYU and Meta, it utilizes a specialized data collection tool called 'The Stick'—constructed from affordable components like a reacher-grabber and an iPhone—to record human demonstrations. By leveraging the 'Homes of New York' (HoNY) dataset and Home Pretrained Representations (HPR), Dobb-E allows users to teach a mobile robot (specifically the Hello Robot Stretch) new household tasks, such as opening drawers or turning off lights, in approximately 20 minutes (5 minutes of demonstration followed by 15 minutes of fine-tuning).

Key capabilities

The Stick: A low-cost demonstration tool using an iPhone and 3D-printed parts.
Home Pretrained Representations (HPR): A vision model pretrained on diverse home environments.
Rapid Learning: Teach a new robotic task in just 20 minutes.
HoNY Dataset: Access to 13 hours of real-world household interaction data.
Open Source: Fully accessible software stack, models, and hardware designs.
Real-world Robustness: Designed to handle household challenges like shadows and variable lighting.
Compatibility: Specifically optimized for the Hello Robot Stretch platform.
Behavior Cloning: Uses imitation learning to map visual inputs to robotic actions.

Pricing

Free — plans below

Open Source

Most popular

0

  • Free access to GitHub repository
  • Pretrained HPR models on Hugging Face
  • 3D printable files for hardware
  • Comprehensive documentation
  • Access to HoNY dataset

Reviews

Be the first to review Dobb-E

No reviews yet

Be the first to share your experience with Dobb-E.

Featured Tools

Handpicked by our team of experts

Suggest an Update

Found outdated information? Help us keep this listing accurate.

FAQ

Dobb-E FAQs

Common questions about Dobb-E