OpenCodePapers

3d-dense-captioning-on-scanrefer-dataset

Image Captioning3D dense captioning
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodeCIDErBLEU-4METEORROUGE-LModelNameReleaseDate
3D CoCa: Contrastive Learners are 3D Captioners✓ Link85.4245.5630.9561.983D CoCa2025-04-13
See It All: Contextualized Late Aggregation for 3D Dense Captioning83.1442.1727.9259.44See It All2024-08-14
Bi-directional Contextual Attention for 3D Dense Captioning80.1440.1627.7656.10BiCA2024-08-13
Vote2Cap-DETR++: Decoupling Localization and Describing for End-to-End 3D Dense Captioning✓ Link76.3641.3728.7060.00Vote2Cap-DETR++2023-09-06
End-to-End 3D Dense Captioning with Vote2Cap-DETR✓ Link71.4539.3428.2559.33Vote2Cap-DETR2023-01-06
3DJCG: A Unified Framework for Joint Dense Captioning and Visual Grounding on 3D Point Clouds60.8639.6727.4559.023DJCG2022-01-01
MORE: Multi-Order RElation Mining for Dense Captioning in 3D Scenes✓ Link58.8935.4126.3655.41MORE2022-03-10
Spatiality-guided Transformer for 3D Dense Captioning on Point Clouds✓ Link58.0635.3026.1655.03SpaCap3d2022-04-22
Scan2Cap: Context-aware Dense Captioning in RGB-D Scans53.7334.2526.1454.95Scan2Cap2020-12-03
Contextual Modeling for 3D Dense Captioning on Point Clouds50.2926.6422.5744.71Contextual2022-10-08
Context-Aware Alignment and Mutual Masking for 3D-Language Pre-Training✓ Link50.0231.8724.5351.173D-VLP2023-01-01
X-Trans2Cap: Cross-Modal Knowledge Transfer using Transformer for 3D Dense Captioning✓ Link41.5223.8321.9044.97χ-Tran2Cap2022-03-02