Lune

ICLR2025Top-tier venue

Uni2Det: Unified and Universal Framework for Prompt-Guided Multi-dataset 3D Detection

Yubin Wang, Zhikang Zou, Xiaoqing Ye, Xiao Tan, Errui Ding, Cairong Zhao

Abstract

We present Uni 2 Det, a brand new framework for unified and universal multi-dataset training on 3D detection, enabling robust performance across diverse domains and generalization to unseen domains. Due to substantial disparities in data distribution and variations in taxonomy across diverse domains, training such a detector by simply merging datasets poses a significant challenge. Motivated by this observation, we introduce multi-stage prompting modules for multi-dataset 3D detection, which leverages prompts based on the characteristics of corresponding datasets to mitigate existing differences. This elegant design facilitates seamless plug-andplay integration within various advanced 3D detection frameworks in a unified manner, while also allowing straightforward adaptation for universal applicability across datasets. Experiments are conducted across multiple dataset consolidation scenarios involving KITTI, Waymo, and nuScenes, demonstrating that our Uni 2 Det outperforms existing methods by a large margin in multi-dataset training. Notably, results on zero-shot cross-dataset transfer validate the generalization capability of our proposed method.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

3 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Towards Large-Scale 3D Representation Learning with Multi-Dataset Point Prompt Training cites Segment Anything
  • Towards Large-Scale 3D Representation Learning with Multi-Dataset Point Prompt Training cites Simple Multi-dataset Detection
  • COME: Dual Structure-Semantic Learning with Collaborative MOE for Universal Lesion Detection Across Heterogeneous Ultrasound Datasets cites Segment Anything
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers3

  1. Hierarchical Cross-Modal Prompt Learning for Vision-Language ModelsHao Zheng, Shunzhi Yang, Zhuoxin He, Jinfeng Yang et al.ICCV 2025 · 5 citations
  2. COME: Dual Structure-Semantic Learning with Collaborative MOE for Universal Lesion Detection Across Heterogeneous Ultrasound DatasetsLingyu Chen, Yawen Zeng, Yue Wang, Peng Wan et al.ICCV 2025 · 1 citation
  3. Adversarial Domain Prompt Tuning and Generation for Single Domain GeneralizationZhipeng Xu, De Cheng, Xinyang Jiang, Nannan Wang et al.CVPR 2025

Builds on18

  1. Segment AnythingAlexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao et al.ICCV 2023 · 13,211 citations
  2. Voxel R-CNN: Towards High Performance Voxel-based 3D Object DetectionJiajun Deng, Shaoshuai Shi, Peiwei Li, Wengang Zhou et al.AAAI 2021 · 1,128 citations
  3. Simple Multi-dataset DetectionXingyi Zhou, Vladlen Koltun, Philipp KrähenbühlCVPR 2022 · 85 citations
  4. Towards Large-Scale 3D Representation Learning with Multi-Dataset Point Prompt TrainingXiaoyang Wu, Zhuotao Tian, Xin Wen, Bohao Peng et al.CVPR 2024 · 39 citations
  5. mDALU: Multi-Source Domain Adaptation and Label Unification with Partial DatasetsRui Gong, Dengxin Dai, Yuhua Chen, Wen Li et al.ICCV 2021 · 27 citations

Related papers

  1. Towards Universal LiDAR-Based 3D Object Detection by Multi-Domain Knowledge TransferGuile Wu, Tongtong Cao, Bingbing Liu, Xingxin Chen et al.ICCV 2023 · 7 citations
  2. CMT: Co-training Mean-Teacher for Unsupervised Domain Adaptation on 3D Object DetectionShijie Chen, Junbao Zhuo, Xin Li, Haizhuang Liu et al.ACM MM 2024 · 5 citations
  3. Unifying Voxel-based Representation with Transformer for 3D Object DetectionYanwei Li, Yilun Chen, Xiaojuan Qi, Zeming Li et al.NeurIPS 2022 · 401 citations
  4. Unified Domain Generalization and Adaptation for Multi-View 3D Object DetectionGyusam Chang, Jiwon Lee, Donghyun Kim, Jinkyu Kim et al.NeurIPS 2024 · 19 citations
  5. One for All: Multi-Domain Joint Training for Point Cloud Based 3D Object DetectionZhenyu Wang, Yali Li, Hengshuang Zhao, Shengjin WangNeurIPS 2024 · 13 citations
Dusk over the sea between two cliffs drawn in fine vertical lines