Submission: free6d(refine)/HB/free6d-refine-hb

Download submission
Submission name free6d-refine-hb
Submission time (UTC) Nov. 12, 2026, 10:27 a.m.
User jzhao12
Task Model-based 6D localization of unseen objects
Dataset HB
Description
Evaluation scores
AR:0.646
AR_MSPD:0.713
AR_MSSD:0.628
AR_VSD:0.598
average_time_per_image:2.600

Method: free6d(refine)

User jzhao12
Publication
Implementation https://github.com/JZhao12/free6d
Training image modalities None
Test image modalities RGB
Description

Submitted to: BOP Challenge 2025, model-based 2D detection and 2D segmentation of unseen objects, BOP-Classic-Core (LM-O, T-LESS, TUD-L, IC-BIN, ITODD, HB, YCB-V), BOP'19 test targets.

Training data: none. Free6D is zero-shot: SAM 2.1 Hiera-L and DINOv3 ViT-L/16 are used with their public pretrained weights (sam2.1_hiera_large.pt, dinov3_vitl16_pretrain_lvd1689m), no fine-tuning on BOP data.

Onboarding data: 42 templates per object rendered with pyrender from the CAD model (viewpoints of the level-0 icosphere, camera distance 0.4 of the predefined poses, black background, uniform grey material for T-LESS). No real onboarding images.

Used 3D models: the default BOP CAD models (models/, models_cad/ for T-LESS).

Computer specifications Intel Core Ultra 9 285K, 24 cores; NVIDIA GeForce RTX 5070, 12 GB; Ubuntu 24.04.5 LTS, Linux 7.0.0-42-generic