FPv1

Dataset Information
Modalities
3D, Point cloud
Introduced
2022
License
Unknown
Homepage
Official Website
Contents

Overview

FPv1 (prior name FAUST-partial) is a 3D registration benchmark dataset created to address the lack of data variability in the existing 3D registration benchmarks such as: 3DMatch, ETH, KITTI.

The original FAUST training dataset is comprised of 100 3D scans of human bodies.

The benchmark generation for a single scan from the FAUST training dataset can be summarized as follows:

  1. Make xz-plane the floor by translating the minimal bounding box point of the scan to the origin
  2. Surround the scan with a regular icosahaedron. Each point of the icosahaedron acts as a viewpoint
  3. For each viewpoint, create a partial point cloud using the hidden point removal algorithm

Finally, for a pair of partial point clouds with the desired overalp, generate a random rotation from the desired rotation range and translation range.

Variants: FPv1, FAUST-partial (60%+ overlap, Rot 0-45, Trans -50-50), FAUST-partial (60%+ overlap, Rot 0-45, Trans -50-50, trained on 3DMatch)

Associated Benchmarks

This dataset is used in 1 benchmark:

Recent Benchmark Submissions

Task Model Paper Date
Point Cloud Registration Greedy Grid Search Challenging the Universal Representation of … 2022-11-29
Point Cloud Registration GeoTransformer Geometric Transformer for Fast and … 2022-02-14
Point Cloud Registration FCGF + YOHO-C You Only Hypothesize Once: Point … 2021-09-01
Point Cloud Registration FCGF + YOHO-O You Only Hypothesize Once: Point … 2021-09-01
Point Cloud Registration FCGF + PointDSC PointDSC: Robust Point Cloud Registration … 2021-03-09
Point Cloud Registration SpinNet SpinNet: Learning a General Surface … 2020-11-24
Point Cloud Registration DIP Distinctive 3D local deep descriptors 2020-09-01

Research Papers

Recent papers with results on this dataset:

External Links: