3D Reconstruction

The computer vision task of recovering the 3D structure of a scene - geometry, camera positions, and sometimes object trajectories - from one or more 2D images or video frames. Techniques range from classic structure-from-motion and multi-view stereo to modern learning-based methods like neural radiance fields (NeRF), Gaussian splatting, and transformer models such as VGGT (Visual Geometry Grounded Transformer). In accessibility, 3D reconstruction enables spatial audio that is anchored to the actual scene rather than the camera frame, which is critical for blind and low-vision users because screen-space spatial audio produces disorienting jumps whenever the camera cuts or pans. It also underpins indoor navigation aids, AR product exploration, and sensory substitution systems that need a consistent model of the surrounding space.