Datasets
Standard Dataset
Three-dimensional object recognition and segmentation in cluttered scenes
- Citation Author(s):
-
AjmalMian
- Submitted by:
- Ajmal Mian
- Last updated:
- Fri, 01/13/2023 - 04:29
- DOI:
- 10.21227/6rgk-8697
- Data Format:
- Research Article Link:
- License:
493 Views- Categories:
- Keywords:
0 ratings - Please login to submit your rating.
Abstract
This dataset contains 3D models of 5 objects and numerous scenes where the objects are placed randomly creating occlusions and cluttered scenes. The 3D models are to be used to find them in the scene (object recognition) and segment them as well.
Instructions:
Ground truth locations of the objects are given in the GroundTruth_3Dscenes.zip file. Using Scanalyze (Stanford University), you can verify the .xf files.
Funding Agency:
Australian Research Council
Grant Number:
DP0344338
Dataset Files
- This file contains the 3D models and 50 scenes. The models folder contains multiple 2.5D views as well as the full 3D models. models_and_scenes.zip (177.13 MB)
- The location (rotation and translation) of each object in the 50 scenes. GroundTruth_3Dscenes.zip (51.32 kB)
Comments
dd
hi