Welcome to the first post in our monthly blog series where we bring you up to speed on recent happenings in the FiftyOne community and celebrate noteworthy milestones. 🙌 🚀
Voxel51’s Commitment to Open Source and Community
If you’re new to Voxel51, open source, transparency, and giving back to the computer vision community are what we are all about! Whether it’s developing the open source FiftyOne computer vision toolset to help engineers and data scientists build high-quality datasets and models, sponsoring Meetups to help members boost their computer vision knowledge, or giving to charitable causes on behalf of the community, Voxel51 is committed to bringing transparency and clarity to the world’s data.
Community Spotlights
We love hearing how FiftyOne helps you solve challenges and reach new heights! Curious what sorts of use cases are possible with FiftyOne? Here are just a few highlights from what people in the community have to say.
Research and Development – Raytheon
The Raytheon Technologies Research Center serves as the company’s innovation hub. RTRC’s engineers, scientists, and researchers anticipate the discoveries destined to change everything, and they transform that research into the solutions and products that help the company’s businesses shape the future.
“We use FiftyOne to organize large research datasets. My favorite feature is the ability to view distributions over image attributes in the dataset, and filter the dataset by those attributes.”
— Brett Israelsen, Principal Research Scientist
Agriculture – Taranis
Taranis is a leading precision agriculture intelligence platform that uses sophisticated computer vision, data science, and deep learning algorithms to effectively monitor fields. Taranis offers a full-stack solution for high precision aerial surveillance imagery to prevent crop yield loss due to insects, diseases, and weeds.
“We’ve been using FiftyOne for over a year and it has drastically changed the way we work. The ability to easily display and analyze our images and their metadata, including experiment results, has been a refreshing change compared to the way we’ve worked before – mainly writing our own metrics and viewers. I’ve personally used FiftyOne for a segmentation model I’ve trained – trying to analyze the results and visually see what my model outputs has been really easy and fluid thanks to FiftyOne.”
— Ido Greenfeld, AI Team Lead
Waste Management/Materials Recovery – Binit
Binit is the first fully integrated analytics and data platform for waste management, helping Materials Recovery Facilities analyze materials and boost revenues via data.
“FiftyOne is the backbone of our Data Engine. It helps us to clean and relabel our datasets efficiently, and convert our model predictions into large scale training datasets with a little help from CVAT.”
— George Pearse, Machine Learning Engineer
FiftyOne Community Rewards
Is your organization using FiftyOne to solve interesting computer vision problems? Share your success story and claim a box of community rewards as a thank you!
FiftyOne 0.19 Is Here!
FiftyOne adoption continues to accelerate with more than 590k downloads to date! Here’s the latest news on the product side: FiftyOne 0.19 is here and it’s packed with a bunch of new features including an all-new customizable Spaces framework, in-App embeddings visualization, saved views, on-disk segmentations, new UI filtering options, and more!
Check out the new features in the announcement blog post, the latest release notes, and in the live demo & AMA with Voxel51 CTO Brian Moore on February 28 @ 10 AM PT. Oh yea, you can also see it for yourself! It’s easy to get up and running in just a few minutes.
FiftyOne on GitHub
GitHub is home to the open source FiftyOne project. Here’s the latest snapshot of what’s happening in the FiftyOne GitHub repo:
- Total stars: 2,562
- Total contributors: 56 – Shout out to recent contributors shortcipher3, nmanovic, jwertherUM and kalpit-S!
- Total used by: 245 repositories
- Total forks: 306
- Total issues closed so far this year: 185
- Total commits so far this year: 392
FiftyOne Community Slack
The FiftyOne Community Slack channel is where you can join more than 1,360 machine learning engineers and data scientists using FiftyOne to improve the quality of their computer vision data and build better models. Ask questions, answer questions, or simply follow along with the discussion!
To make it easy to catch the highlights, every Friday we recap interesting questions and answers from Slack in Tips & Tricks blog series. Recent posts include:
- Adding & Merging Data Tips & Tricks – Feb 17
- Tips & Tricks – Feb 10
- Model Evaluation Tips & Tricks – Feb 3
- Tips & Tricks – Jan 27
- View Stages Tips & Tricks – Jan 20
Computer Vision Meetups
Voxel51 sponsors 13 virtual Computer Vision Meetups around the world. (To join, visit the Meetup link and scroll down to find the location friendliest to your time zone.)
The Computer Vision Meetups are geared towards data scientists, machine learning engineers, and open source enthusiasts who want to expand their knowledge of computer vision and complementary technologies. We put an emphasis on open source software, and speakers who are computer vision practitioners or academics doing research in the field.
Recapping the Feb ‘23 Meetup
We recently held the February ‘23 Computer Vision Meetup that showcased these topics and speakers:
- Breaking the Bottleneck of AI Deployment at the Edge with OpenVINO — Paula Ramos, PhD (Intel)
- Understanding Speech Recognition with OpenAI’s Whisper Model — Vishal Rajput (AI-Vision Engineer)
Get the Meetup recap, including video playbacks, executive summaries, slides, and Q&A.
Upcoming: March ‘23 Meetup
Now that the event has passed, check out the Meetup Recap to tune into these talks:
- Lightning Talk: A Recycling Max Pooling Module for 3D Point Cloud Analysis — Jiajing Chen (PhD Candidate, Syracuse University)
- Lighting up Images in the Deep Learning Era — Soumik Rakshit, ML Engineer (Weights & Biases)
- Taking Computer Vision Models in Notebooks to Production — Sumanth P (ML Engineer)
Event Calendar
Check out these upcoming FiftyOne and computer vision events.
- Feb 28: FiftyOne 0.19 Live Demo & AMA
- Mar 9: March Computer Vision Meetup
- Mar 22, NVIDIA GTC: Creating Digital Twins & Simulations of Industrial Robotic Workcells for Smart Factories, RIOS
- Apr 13: April Computer Vision Meetup
- May 11: May Computer Vision Meetup
Community-Powered Charitable Contributions
In the spirit of giving and in response to global circumstances over the last few years, we made some enhancements to our event strategy. Two notable changes: we added more virtual events so that more people can participate without the requirement to travel; and we applied our swag budget to donate to charitable causes on behalf of the community.
During each of our virtual Meetups, we ask attendees to vote for their favorite charity, and then we make a donation to the charity that receives the most votes.
In the last four months, on behalf of the computer vision community, we’ve made donations to these admirable causes: Children International, Foundation Fighting Blindness, Turkey-Syria Earthquake Relief by Direct Relief, and World Literacy Foundation.
New Docs, Blogs, Videos, and Tutorials
We want everyone to be successful with FiftyOne, and one of the ways we try to do that is by publishing resources that you might find helpful and handy. Here’s a list of some of the new documentation, blogs, videos, tutorials, integrations, and cheat sheets that you may want to check out.
Blogs
- Automatically Set Up a New ML Project, Pain Free
- How Computer Vision is Changing Agriculture in 2023
- People @ Voxel51: Spotlight on Jimmy Guerrero
- The Making of Avatar: The Way of Water
- Recapping the Computer Vision Meetup – January 2023
- Finding Images with Words
- Exploring the Berkeley Deep Drive Autonomous Vehicle Dataset
Videos
- Computer Vision Meetup: AI Deployment at the Edge with OpenVINO
- Computer Vision Meetup: Speech Recognition Using OpenAI’s Whisper Model
- Exploring the Berkeley Deep Drive Autonomous Vehicle Dataset
- Computer Vision Meetup: Intro to Hugging Face Transformers
- Computer Vision Meetup: Hyperparameter Scheduling
- Pandas-Style Queries for Computer Vision Data
Tutorials and Cheat Sheets
- Nearest Neighbor Embeddings Classification with Qdrant
- Exploring Open Images V7 Tutorial
- Performing pandas-style queries in FiftyOne Tutorial
- pandas vs FiftyOne Cheat Sheet
- FiftyOne Terminology Cheat Sheet
- Views Cheat Sheet
- Filtering Cheat Sheet
New & Updated Documentation
We published a bunch of new docs to help you make the most of your FiftyOne experience:
- Using custom plugins
- Working with sidebar modes
- Customizing the sidebar layout with sidebar groups
- Viewing field-level descriptions with a field tooltip
- Declaring custom dynamic attributes on datasets
- Storing field-level metadata on datasets
- Option to import annotation IDs when loading data stored in COCO format
- Including the export directory in the dataset.yaml file generated by YOLOv5 exports
- Using CUDA devices when running the CLIP model from the Model Zoo
And FiftyOne Teams documentation is now publicly available!
We Are Hiring!
Voxel51 is on a mission to bring transparency and clarity to the world’s data. We’re growing quickly and are looking for people to grow with us. Voxel51 team members are fueled by learning, adapt quickly to face new challenges, and aim to shatter the status quo.
Here’s a sample of some of our open, remote positions:
- Account Executive
- DevOps Engineer
- VP of Engineering
