As someone who has used this service, the ease of creating and controlling an organized annotated dataset was the most valuable characteristic. The user interface is extremely easy to navigate and made organizing labeling work significantly less time-consuming, which is important. This service has very useful collaborative features for communicating with other team members and for annotators to look at the annotations they had made. There is also easy integration with other systems and services that improved the workflow. Review collected by and hosted on G2.com.

At times, the more advanced features could benefit from a bit of extra configuration to utilize to their fullest value. Additionally, it would be beneficial to have more options outlined for built-in automations to speed up the annotation process. Nothing serious, particularly considering the quality of service provided. Review collected by and hosted on G2.com.

I love how it is an all in one platform and I can do everything from labelling to reviewing my data all in one place. It's fast and great for academic use as well. Review collected by and hosted on G2.com.

Some of the interface feels slightly outdated and it is missing some of the newer features that platforms like Roboflow have. Review collected by and hosted on G2.com.

The best feature of Segments.ai is its specialization and efficiency in multi-sensor data labeling, which is critical for autonomous systems and robotics. Review collected by and hosted on G2.com.

Some users have noted that annotating complex, multi-sensor data (like 3D point clouds combined with 2D images) requires a bit of a learning curve to master the interface and tools. Review collected by and hosted on G2.com.

The tool has many features for segmentation annotation. It makes the process seamlessly easy compared to SAM based model annotations. Even with low resolution images it works amazing. Out all the tools I have tried this was faster. On top they have Academic license. I got the license within few minutes and started annotation the data. The response from their team is faster than I expected. Review collected by and hosted on G2.com.

I find the multi sensor annotation required a bit of learning curve. Just like in semantics if they have some semi-automatic it would be top annotation platform. Review collected by and hosted on G2.com.

I am a master's student, and I couldn't find a good tool for 3D point cloud annotation until I found Segments.ai. They generously provided me with a free academic license. This tool was exactly what I needed. It is easy to use, accepts general formats like PCD (not just BIN formats), and they send helpful emails and Google Colab notebooks. Review collected by and hosted on G2.com.

There isn't much to dislike about Segments.ai. I wish they made more videos on YouTube as they are very helpful. Review collected by and hosted on G2.com.

The best thing about segments.ai is how fast one can label images. With other tools that I have used it could take up to 5 minutes to label one image. But with segments.ai it can take as little as 10-20 seconds to label one image with several classes.

The reason for why it is so easy to label images with segments.ai is that this product comes with several different features such as edit with superpixels, brush, polygon etc and it also comes with autosegmentation (which as of now is in beta mode).

Moreover, as a manager it is also very easy to see statistics for each labeler and this product also has a very nice review function so that you easily can correct mistakes. However, mistakes seldomly happen, since this tool makes it so easy to label images.

Furthermore, once you are done with a labeling project there is an amazing python SDK that makes it very easy to download both images and segmentation bitmaps.

I would also like to add that when I encounter problems with segments.ai (which rarely happens), their team is very quick to respond and they are extremely helpful. Review collected by and hosted on G2.com.

So far I have not encountered anything that I dislike about segments.ai Review collected by and hosted on G2.com.

An easy interface and export function. Free academic access. Review collected by and hosted on G2.com.

It would be nice to have options to export annotations straight from the web version, without using Pythin SDA. Review collected by and hosted on G2.com.

I love all its features and how easy it is for both 2D and 3D labelling. The dataset creation and export is simple and intuitive. Labelling, especially in 2D, relies on many tools (superpixels, brushes, etc) that make the experience easy and straightforward to explain to an inexperienced person. Review collected by and hosted on G2.com.

The 3D visualization gets stuck at times because it's very computationally heavy. It would be cool to introduce a feature for "hierarchical labeling", meaning to have multiple instance ids for the same pixel/point. This is useful, for example, in the agricultural domain where both plant instance and leaf instance segmentation are useful, thus each pixel has to be associated to a specific plant and to a specific leaf. Review collected by and hosted on G2.com.

Versatile Annotation Options: The diverse range of annotation types, from bounding boxes to keypoints, empowers me to accurately label various types of data, ensuring comprehensive annotation for my projects.

AI-Assisted Labeling: The AI-powered pre-labeling feature is a true time-saver. It not only speeds up the annotation process but also enhances accuracy by intelligently predicting labels, elevating the quality of my annotated datasets. Review collected by and hosted on G2.com.

Complexity for New Users: Data annotation tools can sometimes have a learning curve, especially for users who are new to the concept of data labeling or have limited technical expertise. Navigating the platform, understanding annotation types, and using advanced features might be challenging initially.

Quality Control: Ensuring the accuracy and consistency of annotations can be a challenge, especially when dealing with complex or large datasets. Users might face difficulties in maintaining high-quality annotations across different annotators or addressing discrepancies in labeling. Review collected by and hosted on G2.com.

It is easy to work together on projects. The UI is easy to understand and the API is fast. Review collected by and hosted on G2.com.

Could be faster on large point cloud sequence Review collected by and hosted on G2.com.