Camera Configuration

In the "Cameras" tab, users can configure various camera modalities to attach to their robot. These camera settings are essential for visualizing the simulation and gathering visual data for processing by AI models.

Main Interface Overview

Configurable Camera Modalities

The platform allows configuration of several camera modalities, each serving a unique purpose within the simulation:

  • RGB: Captures standard 3-channel color images, as seen by the human eye.

  • Depth Perspective: Provides depth information from the camera's perspective, which is useful for 3D reconstruction and understanding the relative positions of objects.

  • Depth Planar: Similar to Depth Perspective but provides depth information on a planar (2D) scale.

  • Depth Vis: Visualizes depth data for easier interpretation by users.

  • Disparity: Offers a measure of the difference in image location of an object seen by left and right cameras, which is related to the distance of the object from the cameras.

  • Segmentation: Used for image segmentation; it labels each pixel of the image according to the category of the object it belongs to.

  • Surface Normals: Provides information about the orientation of surfaces relative to the camera's viewpoint.

  • Infrared Camera: Simulates a non-radiometric infrared camera where objects can be assigned different pixel values based on their temperature.

  • Optical Flow: Measures the motion of objects relative to the camera, useful for determining the speed and direction of moving objects.

Common Camera Parameters

Each camera modality can be configured with the following common parameters:

  • Width: The width of the camera's image in pixels.

  • Height: The height of the camera's image in pixels.

  • Field of View (deg): The extent of the world that is observable at any given moment by the camera, in degrees (horizontal field of view).

Tip

A wider field of view allows the camera to capture more of the environment but can also introduce more distortion into the images. Adjust this setting based on the requirements of your simulation.

Warning

High-resolution cameras may significantly increase the computational load of the simulation, reducing the FPS of the simulation.

After finalizing the camera configurations, proceed to the 'Sensors' tab to configure additional sensors before moving on to 'AI Models' for setting up the intelligence and processing algorithms.