Ray-ONet: Efficient 3D Reconstruction From A Single RGB Image
This addresses efficient 3D modeling for computer vision applications, offering a significant speed improvement over prior methods.
The paper tackles 3D reconstruction from single RGB images by proposing Ray-ONet, which predicts occupancy probabilities along rays to improve accuracy over Occupancy Networks while reducing inference complexity to O(N²). It achieves state-of-the-art performance on ShapeNet with over 20× speed-up at 128³ resolution and similar memory footprint.
We propose Ray-ONet to reconstruct detailed 3D models from monocular images efficiently. By predicting a series of occupancy probabilities along a ray that is back-projected from a pixel in the camera coordinate, our method Ray-ONet improves the reconstruction accuracy in comparison with Occupancy Networks (ONet), while reducing the network inference complexity to O($N^2$). As a result, Ray-ONet achieves state-of-the-art performance on the ShapeNet benchmark with more than 20$\times$ speed-up at $128^3$ resolution and maintains a similar memory footprint during inference.