Dual-Modality AI Perception Framework
This technology adopts a 720P stereo video and depth camera, using a single RGB image stream as the input for AI inference while simultaneously generating a high-precision depth map. This enables native alignment between semantic information and spatial distance data. Through a single capture process, it can concurrently perform object recognition, spatial localization, and real-time obstacle avoidance, significantly simplifying system architecture and data fusion workflows.