Core
Technologies

B.Box Bounding Box is a fundamental technology for object detection. It involves drawing precise rectangular boxes around objects in images or videos, allowing AI to classify and identify an object's presence and location.
Ex : Cars, pedestrians, traffic lights, products displayed in a store, customers.

Polyline Suitable for defining linear topographical features like road lanes, building outlines, or public utility lines, Polyline connects multiple points to create accurate lines.
Ex : Traffic lanes, center lines, crosswalk lines, boundaries of fields in aerial imagery, river courses.

Polygon For accurate pixel-level object segmentation, Polygon uses multi-point shapes to meticulously trace the complex and irregular contours of objects. This data enables AI to precisely understand the exact shape and area of an object and accurately separate it from the background.
Ex : Tumors in medical images, precise segmentation of specific organs, accurate differentiation of topographical features like buildings, forests, or puddles in satellite imagery.

Keypoint Keypoint is used for pose estimation and precise feature detection. It accurately identifies specific anatomical or structural landmarks.
Ex : Estimating posture by identifying the location of joints like shoulders, elbows, knees, and ankles, or detecting facial features such as eyes, nose, and mouth through facial recognition.

3D Cuboid This technology is essential for 3D object detection and spatial understanding in real-world environments. It labels objects with 3D bounding boxes that define their length, width, height, and orientation in three-dimensional space. This allows AI in autonomous systems and robotics to accurately perceive depth and spatial relationships.
Ex : Autonomous vehicles using 3D recognition of surrounding cars, buildings, and signs to predict collision risks, or robots understanding the precise location and volume of objects in a workspace for manipulation.