Submission: FoundPose+FeatRef/HB/FoundPose+FeatRef

Download submission
Submission name FoundPose+FeatRef
Submission time (UTC) Nov. 15, 2023, 6:14 p.m.
User epi
Task Model-based 6D localization of unseen objects
Dataset HB
Training model type Default
Training image type None
Description
Evaluation scores
AR:0.585
AR_MSPD:0.697
AR_MSSD:0.546
AR_VSD:0.511
average_time_per_image:2.820

Method: FoundPose+FeatRef

User epi
Publication
Implementation
Training image modalities RGB
Test image modalities RGB
Description

The presented results were achieved by FoundPose with the featuremetric refinement (row 6 of Table 1 in [A]).

In this submission, FoundPose uses default CNOS-FastSAM [B] segmentations provided for BOP'23. For pose estimation, the method uses features from layer 18 of DINOv2 (ViT-L) with registers [C].

Note that FoundPose doesn't do any task-specific training -- it only uses frozen FastSAM (via CNOS) and frozen DINOv2.

[A] Anonymous: FoundPose: Unseen Object Pose Estimation with Foundation Features.

[B] Nguyen et al.: CNOS: A Strong Baseline for CAD-based Novel Object Segmentation, ICCVW 2023.

[C] Darcet et al.: Vision transformers need registers, arXiv 2023.

Computer specifications Tesla P100 16GB