Supplementary Material for Learning 3D Shape Completion from Laser Scan Data with Weak Supervision
نویسندگان
چکیده
This document provides additional details regarding the used datasets and further experimental results complementary to the main paper. In Section 2, we discuss technical details regarding the introduced, synthetic datasets derived from ShapeNet [2], referred to as SN-clean and SN-noisy, the synthetic dataset derived from ModelNet [17], as well as the dataset extracted from KITTI [9]. Then, in Section 3, we provide additional details concerning architecture followed by a discussion of the training procedure in Section 4. In Section 5, we discuss our implementation of the mesh-to-mesh distance used for evaluation. Subsequently, in Section 6 we discuss the evaluation of the approach by Engelmann et al. [7] in more detail and, finally, present additional experiments in Section 7.
منابع مشابه
Learning 3D Shape Completion from Laser Scan Data with Weak Supervision
3D shape completion from partial point clouds is a fundamental problem in computer vision and computer graphics. Recent approaches can be characterized as either datadriven or learning-based. Data-driven approaches rely on a shape model whose parameters are optimized to fit the observations. Learning-based approaches, in contrast, avoid the expensive optimization step and instead directly predi...
متن کاملDeep Supervision with Shape Concepts for Occlusion-Aware 3D Object Parsing – Supplementary Material
In this supplementary material, Section 2 details the 3D annotation for CAD models and real images as well as our approach to compute the car yaw angle given a 3D skeleton. Next, in Section 3, we present the instance segmentation algorithm in detail for our experiments on PASCAL3D+. Finally, we provide more quantitative results on KITTI-3D in Section 4 and demonstrate more qualitative results o...
متن کاملHigh Resolution Shape Completion Using Deep Neural Networks for Global Structure and Local Geometry Inference (Supplementary Material)
متن کامل
Combining Texture and Shape Cues for Object Recognition with Minimal Supervision
We present a novel approach to object classification and detection which requires minimal supervision and which combines visual texture cues and shape information learned from freely available unlabeled web search results. The explosion of visual data on the web can potentially make visual examples of almost any object easily accessible via web search. Previous unsupervised methods have utilize...
متن کاملEfficient 3D Room Shape Recovery from a Single Panorama: Supplementary Material
We explain more details of our work entitled Efficient 3D Room Shape Recovery from a Single Panorama in this supplementary material for completeness.
متن کامل