On a sweltering summer day, we need to stay cool, and BasicAI Cloud just gets refreshed!
BasicAI Cloud in a New UI!
Summer's in full swing, and we've got some refreshing news for you!
On July 3rd, BasicAI released a significant update for BasicAI Cloud, upgrading it to version 0.9. We've completely redesigned BasicAI Cloud’s user interface, and the main color for new palette is now the calming BasicAI blue. Now, keep cool and collected while working on your data annotation projects, just like sipping on a frosty beverage under the shade!
Along with the refreshing blue, we've updated our icons to harmonize with the new color scheme, ensuring a cohesive look throughout the platform.
To enhance readability, we've also given the Task module a makeover. Annotation tasks now come in the form of cards.
We believe the redesigned UI will make it easier for Computer Vision engineers and data annotators to get started with the powerful labeling toolset on BasicAI Cloud. So take a moment to unwind, and immerse yourself in the new user experience that awaits you on BasicAI Cloud v0.9.
Quality Assurance Made Easy
You've heard it before: "Data is the new oil." This couldn't be more accurate in the realm of AI/ML. But remember, not all datasets are created equal: "Garbage in, garbage out."
In Computer Vision applications, pristine training data means a top-notch model. Did you know that quality assurance (QA) consumes about 40% of the time spent on data annotation? This crucial stage can't be skipped, as humans are prone to errors.
With our 7+ years of expertise in AI training data, we've discovered that each annotation project has unique QA requirements. So, we've refined those requirements and fused an ultra-accessible and powerful QA module right into your annotation workflow, and, of course, it's configurable!
On BasicAI Cloud, the Quality Assurance isn't just for review mode; it's available in the Dataset and Task tabs too, once you've set up your project-specific QA rules:
Real-time QA empowers data annotators with instantaneous alerts for oversights or missteps that could derail a particular annotation project.
Batch QA can be deployed to automatically scrutinize the entirety of your annotations.
Out of the box, the default configurable QA rules cater to 50% of the requirements for any annotation project. Can't spot the rules you need? Join our Slack community, and we'll incorporate your required rules into your annotation project.
Point Cloud Annotation: Better, Faster, Stronger
We've also been working non-stop to enhance our point cloud / sensor fusion annotation toolset to make it even more powerful and user-friendly for AI developers. You know how vital data from LiDAR sensors is for computer vision model development, especially in autonomous driving scenarios. With BasicAI Cloud v0.9, we've got more fantastic news for you:
Semi-auto calibration for sensor fusion data? You got it!
Track Line view is now supported for consecutive point cloud frame annotations
Set limits on range while running our auto annotation model
Label point cloud data with style using our useful ranging tool
Poly-line and polygon tools ready to rock for Sensor Fusion data annotation & segmentation
And there's so much more to discover!
Now, AI trainers can be easily calibrating sensor fusion data, annotating drivable regions and lane lines in point cloud data, and running pre-labeling models within a specified range. And of course, our latest QA module is here to ensure top-notch quality checks.
Dataset Checked at One Glimpse
"How many data pieces are in my dataset?", "How many annotations have my team members made?", "How many 'Cars' are there among my annotations?"
We know that as an annotation project team leader, you constantly have questions running through your mind. Worry not! BasicAI Cloud v0.9 now gets you covered.
With the new Datasets Overview feature, you'll have all the answers at your fingertips. As you enter the Datasets page, you'll see a concise yet informative snapshot of your annotation progress. This bird's-eye view will provide you with crucial insights, including:
The total number of data pieces in your dataset
The overall count of annotations made by yourself / your team
A clear chart of Classes (say, cars, bicycles, traffic lights, etc.)
This comprehensive Overview ensures that you can always stay on top of your datasets, enabling you to make informed decisions and effectively manage your team's efforts.
Label Data with Style on BasicAI Cloud V0.9. Experience All Cool Annotation Features at zero Cost!