HOME CHALLENGES DATASETS TASKS LEADERBOARDS SUBMIT RESULTS Sign in

Submission: ocdit/TUD-L/dinov3-ocdit-dpt

Download submission
Submission name dinov3-ocdit-dpt
Submission time (UTC) June 19, 2026, 3:50 p.m.
User mulmer
Task Model-based 2D segmentation of unseen objects
Dataset TUD-L
Description
Evaluation scores
AP:0.559
AP50:0.930
AP75:0.578
AP_large:0.548
AP_medium:0.552
AP_small:0.576
AR1:0.592
AR10:0.592
AR100:0.592
AR_large:0.578
AR_medium:0.577
AR_small:0.575
average_time_per_image:-1.000

Method: ocdit

User mulmer
Publication Ulmer, Maximilian, et al. "Conditional Latent Diffusion Models for Zero-Shot Instance Segmentation.", ICCV 2025
Implementation https://github.com/DLR-RM/oc-dit
Training image modalities RGB
Test image modalities RGB
Description
Computer specifications