| Submission name | |||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Submission time (UTC) | April 30, 2026, 11:56 a.m. | ||||||||||||||||||||||||||
| User | andreacaraffa | ||||||||||||||||||||||||||
| Task | Model-based 2D segmentation of seen objects | ||||||||||||||||||||||||||
| Dataset | ITODD | ||||||||||||||||||||||||||
| Description | |||||||||||||||||||||||||||
| Evaluation scores |
|
| User | andreacaraffa |
|---|---|
| Publication | |
| Implementation | |
| Training image modalities | RGB |
| Test image modalities | RGB |
| Description | Training data: real + provided PBR Used 3D models: CAD models for T-LESS, default models for the other datasets. Notes: one network per dataset was trained. The approach builds on SAM3 and incorporates multi-view cues from CAD models to guide segmentation. Authors: temporary anonymity |
| Computer specifications | GPU L40S; CPU AMD EPYC 9474F @ 1.64GHz |