Submission: Vidal-Sensors18/TUD-L

Download submission
Submission name
Submission time (UTC) Oct. 22, 2019, 7:52 a.m.
User jolvid
Task Model-based 6D localization of seen objects
Dataset TUD-L
Training model type Default
Training image type None
Description
Evaluation scores
AR:0.876
AR_MSPD:0.907
AR_MSSD:0.910
AR_VSD:0.811
average_time_per_image:0.486

Method: Vidal-Sensors18

User jolvid
Publication A Method for 6D Pose Estimation of Free-Form Rigid Objects Using Point Pair Features on Range Data, Sensors, 2018, 18.8: 2678.
Implementation
Training image modalities None
Test image modalities D
Description

Implementation of the method described in “A Method for 6D Pose Estimation of Free-Form Rigid Objects Using Point Pair Features on Range Data”. The method combines a normal-based clustering preprocessing step with an improved matching method for Point Pair Features. In addition, it includes an agglomerative hypothesis clustering, an efficient view-dependent re-scoring process and two scene consistency verification steps. A simple non-maximum suppression step has been included in postprocessing to account for multiple instances. The parameters used are the ones described in the paper and are not optimized for speed or performance on these datasets. Due to the straight forward adaptation to VIVO task, provided scores are not consistent/meaningful.

Computer specifications Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz