Also , Please find the below sample video for how to use AI – Assist for an image.
Documentation
Workspace Management
Project Management
Transforming Data Annotation for the Future
What is JThete Annotate?
JThete Annotate is your all-in-one platform for creating, managing, and optimizing data annotation workflows. Designed for teams and individuals, it simplifies the process of labeling data for AI and machine learning projects with precision and ease.
Shortcuts
Admin Workspace
Annotator Project Workspace
Why Choose JThete Annotate?
Intuitive Workspace Management
Organize your projects seamlessly with dedicated workspaces created for your team
Advanced Annotation Tools
Access a variety of tools, including polygons, bounding boxes, segmentation, semantic segmentation, and more, to suit your project needs.
Collaboration Made Easy
Work with your team in real time with collaborative features to speed up the annotation process.
Scalable and Secure
From small datasets to enterprise-scale projects, JThete ensures high performance and data security.
Get Started in 3 Easy Steps
Login or Sign Up
Access your projects and start annotating in seconds
The first time you use Jtheta.ai you’ll need to sign up. To do this go to https://jtheta.ai/ and click on sign-up for free link and create a personal account for your personal workspace.
Jtheta supports Different Login methods(Using EmailID or Gmail SignIn)
On the sign-up page, start by entering your email address. If you often use Google, a Microsoft email account, or other quick sign-in options, we recommend selecting the corresponding method for a seamless experience. For instance, if you’re already signed into a Google account while browsing, simply click “Continue with Google.” Your JThete.ai account will be instantly created, and you’ll be logged in without delay.
Browser
The JThete.ai annotation interface is designed for optimal performance on the latest version of Google Chrome. For the best experience, we strongly recommend using Chrome on a laptop or desktop.
**Different Browser testing are in-progress**
Create Workspace
In JThete Annotate, the workspace is created based on the tier you select (Free, Starter, Pro, or Enterprise). When you sign up for a JThete account, you’ll be prompted to create a workspace. From there, you can begin organizing and managing your projects, start annotating your data, and take advantage of the platform’s powerful tools—all within your designated workspace.
“Visit the Setup a Project page to quickly get started and annotate your data efficiently.”
- Setup a Project
In JThete Annotate, you can add a new workspace by clicking the “New Workspace” button located in the top-right corner. From there, you can manage your active projects according to the plan you’ve selected, ensuring your workspace aligns with your specific needs and resources.
Manage Your Workspace
To manage a workspace, simply click the settings icon for your current workspace. You’ll find this button in the top-left corner of your JTheta Annotate dashboard, giving you easy access to customize and adjust your workspace settings as needed.
You can manage all your workspace needs directly from the workspace settings, allowing you to customize and control every aspect of your workspace efficiently.
User Management
In JThete Annotate, workspace members can be managed via the Users Page in Jlabel dashboard. Users are categorized by their roles. Admins can add users through the User Management feature. Once added, users will receive an email from JThete to set up their password. After completing the setup, users can log in and access their personal account on the JThete dashboard.
To add someone to your workspace in JThete Annotate, click on the “Add User” button, then enter the required fields along with their assigned role.
Note : Admin can add new users(Annotator, Reviewer & another admin). An email will be sent for verification. Adding Admin himself as Annotator AND/OR Reviewer
Once a user is added to the workspace, they will receive an email to set up their password. After the password is set up, the account will be listed in User Management.
User Access
In JThete Annotate, each user role can be designated as either an Annotator or a Reviewer within a project.
When creating a project, you’ll need to specify each users position in the project, assigning them as either an Annotator or a Reviewer.
Admin Role
In JTheta, the Admin role has the authority to manage users, assign annotators and reviewers, and control task distribution within projects. The Admin ensures that annotation and review workflows are properly assigned based on the Round Robin or other distribution methods.
Annotator Role
The Annotator role in JThete Annotate is responsible for labeling and tagging data within a project. Annotators use the platform’s tools to create annotations such as bounding boxes, polygons, or other markers, depending on the project’s requirements. Their primary task is to accurately and efficiently label data to ensure high-quality input for AI and machine learning models.
Reviewer Role
The Reviewer role in JThete Annotate is responsible for quality assurance within a project. Reviewers carefully inspect the annotations created by Annotators to ensure accuracy, consistency, and alignment with project guidelines. They can approve, modify, or reject annotations as needed, playing a critical role in maintaining high-quality data for AI and machine learning models.
Role-Based User Access in JTheta
Below is a table outlining the permissions for Admin, Annotator, and Reviewer in JTheta:
Feature/ActionAdminAnnotatorReviewer
Feature/Action | Admin | Annotator | Reviewer |
|---|---|---|---|
User Management | Can add/remove users | No access | No access |
Project Creation | Can create/edit/delete projects | No access | No access |
Task Assignment | Can assign annotators & reviewers | No access | No access |
Task Distribution Mode | Can configure (Round Robin/Manual) | No access | No access |
View Assigned Tasks | Full access | Can view only their tasks | Can view only their tasks |
Annotate Images | No access | Can create/edit annotations | No access |
Review Annotations | No access | No access | Can approve/reject annotations |
Modify Other Annotations | No access | No access | Can edit before approval |
Export Data | Can export all data | No access | No access |
Analytics Data | Can view project analytics | No access | No access |
Create Dataset
Each project in JThete Annotate must include at least one dataset. After creating a new project, click the “Add Dataset” button to set up a dataset. This dataset will act as the container for your uploaded data. Be sure to take note of the project and dataset IDs for future reference.
When creating a project new images can be added(new dataset) or an existing dataset can be linked to the project.
Our platform currently supports displaying and annotating DICOM, JPEG, PNG, and satellite images in the viewer.
Also, link a dataset while creating a project in Step 2 using an existing dataset already loaded in the admin dashboard.
Project Management
Create a Project
This step-by-step guide will walk you through creating a dataset from scratch using JThete Annotate. You’ll learn how to set up a project, even if you’re new to data annotation.
To create a project in JThete Annotate, ensure you have an account and are logged in. If you haven’t registered yet, sign up here to get started. Once logged in, you’ll need to create a workspace. For detailed instructions on setting up and managing workspaces, visit the Create a Workspace page.
In this guide, we’ll walk you through the process of setting up projects and beginning to annotate your data.
Let’s dive in!
To get started, simply visit JThete Annotate and click on Create a Project.
Jtheta Supports Dashboard views for Admin, Annotator & Reviewer.
Support for Workspace dashboard
Support for Project dashboard
“Visit the Setup a Workspace page to quickly get started and annotate your data efficiently.”
- Setup a Workspace
Once you click the “Create a Project” button, a new project will be created. From there, you can manage, edit, and begin testing data annotation within the project.
Select the Data Modality Type: A prompt will appear asking you to choose the data modality type. Click on the desired option.
Support for Geospatial projects for satellite images
Support YOLOV8m, YOLOV8x & Faster-RCNN models for satellite images
To successfully create a project in JThete Annotate, follow these steps:
Step 1 : Project DetailsProvide the necessary details for your project.
Step 2 : Upload Data – Upload the data you wish to annotate.
Step 3 – Add Classes
Define the annotation classes for your project.Enable Automatic Annotation – Optionally, enable automatic annotation to speed up the process.
Allow Users to Create Classes – Give team members the ability to create their own classes if needed
Step 4 – Add MembersInvite team members to collaborate on the project as annotator and reviewer.
Step 5 – Confirm Project
Review the setup and confirm the project creation.
Once these steps are completed, your project will be ready for annotation.
Round Robin Distribution for Annotators and Reviewers
Overview
Round Robin Distribution ensures that annotation and review tasks are evenly assigned among available users in a cyclic order. This approach helps balance the workload and prevents any single user from being overloaded with tasks.
How It Works
1. Annotator Round Robin Assignment
When new annotation tasks are created, they are assigned to annotators in a sequential manner.
Example (with three annotators: A, B, and C):
Task 1 → Annotator A
Task 2 → Annotator B
Task 3 → Annotator C
Task 4 → Annotator A (cycle repeats)
If an annotator is unavailable (inactive or removed), the system skips them and assigns the task to the next available user.
2. Reviewer Round Robin Assignment
After an annotator completes a task, it is assigned to a reviewer using the same Round Robin method.
Example (with two reviewers: X and Y):
Task 1 (Annotated) → Reviewer X
Task 2 (Annotated) → Reviewer Y
Task 3 (Annotated) → Reviewer X (cycle continues)
If a reviewer is unavailable, the next available reviewer takes over.
Key Features
✅ Fair Workload Distribution – No single user gets an uneven number of tasks.
✅ Automatic Skipping – Inactive or unavailable users are bypassed.
✅ Real-time Adjustments – If a new user joins mid-cycle, they are included in the next rotation.
Scenarios & Edge Cases
If all annotators or reviewers are unavailable, tasks remain unassigned until a user is available.
If an annotator or reviewer leaves a project, the system dynamically redistributes their tasks.
For detailed guidance on creating a project for annotation, please refer to the attached video tutorial. This video demonstrates the step-by-step process, including selecting the data modality and finalizing the project setup.
The Project Settings page allows users to view and modify project details.
Project creator can create class during the project creation process and enable/disable annotators to create new classes or during annotation.
Project creator can enable/disable AI-Assist feature for Annotators
Annotation
Once the project has been created, the annotator can access their dashboard and begin annotating the images. The dashboard provides easy navigation to start the annotation process and manage the data efficiently.
Annotations are saved continuously in the background, so annotators doesn’t lose their work.
“Visit the Setup a Project page to quickly get started and annotate your data efficiently.”
- Setup a Project
Start Annotating
In this section, we’ve created a Road Image Project as a demo. Let’s continue with this project to learn how to annotate data.
To begin the annotation process, simply click the “Annotate” button located in the top-right corner of JThete Annotate.
Learn how to use the different Types Of Annotation of Jtheta Annotate.
- Types of Annotation
When you click the “Annotate” button, you will be directed to the annotation tool in JThete Annotate. Here’s a sample image showing how to use the annotation tool in JThete Annotate. Watch to learn how to annotate your data efficiently.
Search for a class in the list of classes within the class window.
The image editor will display the session and Labeling time.
The annotator dashboard will show annotation status of the images. ToLabel, In-progress, In-review, rework & done
If on inactivity by the annotator/reviewer, the session will timeout.
Users can use zoom sidebar and reset image size(using reset icon) in the image editor
JThete supports several annotation methods, including Bounding box , manual polygon, auto snapping Polygon, semantic segmentation, and instance segmentation, allowing you to annotate images with precision and flexibility.
JTheta supports panning, zooming in, and zooming out on the image canvas while annotating, providing ease of use.
Copy & paste the existing annotation using CTRL-C * CTRL-V
Once you have annotated an image, you can submit it to the reviewer. If you wish to skip an image, simply click the “Skip” button and provide any comments to inform the reviewer.
During annotation, annotator can skip an image with a comment and the Admin can see the annotator’s comment in the project dashboard
Important Points to be remember
Rotate the label in the canvas
Modify and adjust the size of labels on the canvas
Annotate without a review. Annotator annotations are final
Delete the label in the image using Keyboard delete button or delete icon in the objects window
Hide and delete individual annotation using the class/objects window
Pressing ESC button on the canvas will go back from cross hair to pan handle
Hovering over classes/objects in class/objects window will highlight them on the canvas.
Zoom-in and zoom out using fingers(pinch)
When the cursor is moved/hovered over the annotation, the cursor will change to mouse pointer symbol(angled arrow) and a left click will select the annotation to be editable
Also , Please find the below sample video for how to annotate an image
Keyboard Shortcuts – Keyboard shortcuts help minimize the user’s time while annotating.
AI – Assist
JThete Annotate leverages AI-powered assistance to streamline the polygon creation process. With this feature, the AI suggests or auto-generates polygon annotations based on the image content, significantly speeding up the annotation task.
How It Works:
AI Assistance: When annotating an image, the AI automatically detects the key areas and suggests polygon shapes, saving time on manual drawing.
Refining Suggestions: You can refine the suggested polygons, adjust their size, shape, or position, and ensure accuracy.
Increased Efficiency: This feature is particularly useful for large datasets or complex images, helping you annotate faster without compromising on quality.
Enable AI-assisted polygon creation to enhance your annotation workflow in JThete Annotate and make data labeling more efficient.
“Visit the Annotation page to quickly get started and annotate your data efficiently.”
- Annotation
Add Class
In JThete, you can define a class name when creating a project, or you can manually create class labels for each object in an image.
On the annotate page, click Settings to increase the threshold before clicking the AI-Assist button to obtain an accurate output for an image. (Threshold settings for YOLO and Mask-RCNN in the annotation page)
AI-Assist support for BoundingBox, Polygon, Instance segmentation & semantic segmentation
Nested objects as new objects of the classes are created.
Object will be displayed as tree view in Objects window as they are created
After clicking AI-Assist, an “Extra Class Detected” pop-up will appear, prompting you to confirm the objects and choose the desired format (Bounding Box, Polygon, Instance Segmentation, or Semantic Segmentation). Select the format you prefer to proceed.
Review
After the annotator submits the annotated image, it is sent back to the reviewer for evaluation. The reviewer can either accept or deny the image based on the annotator’s response and the quality of the annotations.
During Review, reviewer can reject an image and the image will go back to the annotator for re-work/corrections.
Download Dataset in JSON, CSV, COCO JSON & YOLOV5 formats. The data will be divided 80% for training and 20% for validation.
Once the reviewer completes the review, the annotated file can be exported in various formats, including JSON, CSV, COCO, and YOLOv5.
Types of Annotation
JTheta supports manual bounding box creation, polygon annotation, instance segmentation, and semantic segmentation.
“Visit the Annotation page to quickly get started and annotate your data efficiently.”
- Annotation
Bounding Box
In the context of annotation on the JTheta platform, a bounding box refers to a rectangular annotation tool used to mark and define regions of interest (ROIs) within an image. These regions are often used for labeling objects, text, or features in an image for purposes like:
Object Detection: Identifying specific objects in an image (e.g., cars, people, animals)
How Bounding Boxes Work in JTheta
Drawing the Box: You click and drag on the image to draw a rectangular region around the object or area of interest.
Labeling: Assign a specific label or tag to the bounding box to describe what it contains (e.g., “Car,” “Person”).
Adjustments: Resize or move the box as needed for accuracy.
Bounding boxes are an essential part of annotation workflows, enabling structured and precise data labeling for documentation or AI training purposes. Let me know if you’d like to explore the features in more detail!
Polygon
In the JTheta annotation platform, a polygon is a flexible annotation tool used to outline objects or regions in an image with irregular or non-rectangular shapes. It is especially useful for annotating objects that do not fit neatly into rectangular bounding boxes, such as:
Curved or Complex Shapes: Roads, rivers, or irregularly shaped objects.
Precise Boundaries: Objects like animals, vehicles, or building structures with unique contours.
Fine-Grained Annotation: Areas requiring detailed outlines, like facial features or product components.
How Polygon Works in JTheta
Placing Points:
Click on the image to place points around the object’s perimeter.
Each point becomes a vertex of the polygon.
Connecting Points:
As you place points, they are automatically connected by straight lines to form the polygon’s edges.
Completing the Polygon:
Close the shape by connecting the last point to the first, forming a complete polygon.
Labeling:
Assign a label or category to the polygon to describe the annotated object or region.
Editing:
Adjust points and edges as needed to refine the shape for better accuracy.
Polygons provide higher precision compared to bounding boxes and are commonly used in applications like object segmentation, AI training datasets, and detailed documentation.
Auto Snapping Polygon – Auto-Snapping using SAM
Auto-snapping polygon is a feature in image annotation tools that helps users draw polygons more efficiently by automatically aligning polygon vertices (points) to the nearest relevant edges or boundaries in the image. This feature is particularly useful when annotating objects with well-defined edges or contours, as it reduces the manual effort of precisely placing each vertex and ensures greater accuracy.
Key Features of Auto-Snapping Polygons:
Edge Detection: Automatically detects the edges or contours of objects in the image.
Precision: Vertices snap to the nearest boundary, improving annotation accuracy.
Time-Saving: Reduces the time required for manual adjustments of vertices.
User Control: Often includes toggles to enable or disable snapping based on user preference.
Smoothness: Provides a smoother annotation experience, especially for complex shapes.
This feature is commonly used in machine learning projects for training data preparation, where accurate object boundaries are critical for model performance.
“In jtheta, we provide an Auto-Snapping Polygon Tool feature, as shown in the image below. This tool simplifies the annotation process by automatically generating a polygon around the object with just a click. Once you point to an object and click, the tool detects the object’s edges and creates a polygon, significantly reducing manual effort.”
Semantic Segmentation
Semantic segmentation is a computer vision task that involves dividing an image into regions and assigning a label to every pixel based on the object or category it belongs to. This process creates a pixel-level understanding of the image, where each pixel is classified into predefined categories, such as “car,” “tree,” “road,” or “sky.”
Key Features of Semantic Segmentation:
Pixel-Level Classification: Every pixel in the image is assigned a class label.
Uniform Labeling: All pixels belonging to the same class are grouped under the same label, regardless of whether they are part of separate instances (e.g., all cars are labeled as “car”).
Applications:
Medical Imaging: Segmenting regions like organs or tumors.
Satellite Imaging: Classifying land, water, vegetation, and urban areas.
Example:
In an image of a city street:
The road pixels are labeled as “road.”
The car pixels are labeled as “car.”
The building pixels are labeled as “building.” Semantic segmentation would not differentiate between individual cars but would classify all as “car.”
Instance Segmentation
Instance segmentation is a computer vision task that identifies, delineates, and classifies individual instances of objects in an image. Unlike semantic segmentation, which labels all pixels belonging to a category without differentiating between individual objects, instance segmentation provides both pixel-level masks and instance-specific information for each detected object.
Key Features of Instance Segmentation:
Object Differentiation: Distinguishes between different objects of the same class (e.g., two cars in the same image will have separate masks and labels).
Pixel-Level Accuracy: Provides precise object boundaries for each instance.
Classification and Localization: Combines object detection (bounding boxes) and segmentation to locate and classify objects in the image.
How It Differs from Semantic Segmentation:
Semantic Segmentation: Labels pixels by class (e.g., “car,” “tree,” “road”), without distinguishing between separate objects of the same class.
Instance Segmentation: Identifies individual objects within the same class and provides separate masks for each instance.
Example:
In an image of a parking lot with three cars:
Semantic Segmentation: All car pixels are labeled as “car.”
Instance Segmentation: Each car is assigned a unique identifier (e.g., “car 1,” “car 2,” “car 3”) and has its own mask.
Instance segmentation is a key component of many real-world AI systems, where both the identity and precise shape of objects are critical.
Usage/Billing
In the JTheta workspace dashboard, users can view their workspace usage in the Settings tab, along with their tier-based subscription details.
Workspace Setting showing current usage
Support Tiers based subscription
Store uploaded images in Amazon S3
Create multiple workspaces and projects based on the tier user can choose.
