3D Semantic Segmentation on Rellis-3D 1.0 (test)
62.88mIoUHilDA
Evaluation results
| Method | Links | |
|---|---|---|
| HilDAFine-tuning ratio=10%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 62.88 | |
| CleverDFine-tuning ratio=10%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 60.92 | |
| HilDAFine-tuning ratio=1%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 59.49 | |
| HilDA†Fine-tuning ratio=10%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 59.04 | |
| CleverDFine-tuning ratio=1%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 58.35 | |
| LiMAFine-tuning ratio=10%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 57.15 | |
| HilDA†Fine-tuning ratio=1%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 56.77 | |
| LiMoEFine-tuning ratio=10%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 56.67 | |
| SuperFlowFine-tuning ratio=10%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 55.71 | |
| LiMAFine-tuning ratio=1%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 55.62 | |
| SealFine-tuning ratio=10%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 55.03 | |
| SLidRFine-tuning ratio=10%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 54.57 | |
| LiMoEFine-tuning ratio=1%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 53.74 | |
| RandomFine-tuning ratio=10%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 53.6 | |
| SuperFlowFine-tuning ratio=1%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 52.83 | |
| SealFine-tuning ratio=1%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 51.09 | |
| SLidRFine-tuning ratio=1%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 49.75 | |
| ScaLRFine-tuning ratio=10%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 48.86 | |
| ScaLRFine-tuning ratio=1%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 47.91 | |
| RandomFine-tuning ratio=1%, Backbone=ViT-B/MinkUnet34, Pre-training dataset=nuScenes2026.06 | 38.46 |