J2-9228 — Interim report
1.
Segmentation and reconstruction of 3D models from a point cloud with deep neural networks

The article describes our first attempts on use of neural networks for reconstruction of superquadrics from 3D data.

COBISS.SI-ID: 1537994435
2.
Recovery of superquadrics from range images using deep learning: A preliminary study

The article describes the reconstruction of superquadrics from synthetic and real range images using regression neural networks.

COBISS.SI-ID: 1538269123
3.
Segmentation and recovery of superquadric models using convolutional neural networks

The article describes the problem of segmenting and recovering superquadrics from depht images. First, segmentation is addressed usiing instance segmentation with Mask-RCNNs, which is followed by superquadric-parameter estimation from incomplete data using standard CNNs.

COBISS.SI-ID: 1538520259