mlsanityData to 14 Oct 2026

UniPre3D: Unified Pre-training of 3D Point Cloud Models with Cross-Modal Gaussian Splatting

arXiv 2506.09952 ↗First posted 11 Jun 2025cs.CV cs.AIcited by 6

Not on the map. Why?

Abstract (arXiv)

The scale diversity of point cloud data presents significant challenges in developing unified representation learning techniques for 3D vision. Currently, there are few unified 3D models, and no existing pre-training method is equally effective for both object- and scene-level point clouds. In this paper, we introduce UniPre3D, the first unified pre-training method that can be seamlessly applied to point clouds of any scale and 3D models of any architecture. Our approach predicts Gaussian primitives as the pre-training task and employs differentiable Gaussian splatting to render images, enabling precise pixel-level supervision and end-to-end optimization. To further regulate the complexity of the pre-training task and direct the model's focus toward geometric structures, we integrate 2D features from pre-trained image models to incorporate well-established texture knowledge. We validate the universal effectiveness of our proposed method through extensive experiments across a variety of object- and scene-level tasks, using diverse point cloud models as backbones. Code is available at https://github.com/wangzy22/UniPre3D.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 74 of its 85 references that are in the map. First posted on arXiv 11 Jun 2025.

Builds on

74 of 85 references.

References · 74

  1. 2017Attention Is All You Needcited by 60,361
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  4. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  5. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  6. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  7. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,226
  8. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  9. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  10. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,117
  11. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,897
  12. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,223
38 more
  1. 2020Point Transformercited by 1,470
  2. 20194D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networkscited by 1,208
  3. 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,205
  4. 2023pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstructioncited by 684
  5. 2021Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modelingcited by 653
  6. 2019Deep Hough Voting for 3D Object Detection in Point Cloudscited by 648
  7. 2023Point Transformer V3: Simpler, Faster, Strongercited by 589
  8. 2024MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Imagescited by 566
  9. 2019Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Datacited by 532
  10. 2020PointContrast: Unsupervised Pre-training for 3D Point Cloud Understandingcited by 504
  11. 2022Masked Autoencoders for Point Cloud Self-supervised Learningcited by 459
  12. 2022PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategiescited by 436
  13. 2022Unified-IO: A Unified Model for Vision, Language, and Multi-Modal Taskscited by 421
  14. 2022Point Transformer V2: Grouped Vector Attention and Partition-based Poolingcited by 388
  15. 2022Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Frameworkcited by 379
  16. 2023Splatter Image: Ultra-Fast Single-View 3D Reconstructioncited by 364
  17. 2019DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstructioncited by 333
  18. 2020PointGroup: Dual-Set Point Grouping for 3D Instance Segmentationcited by 321
  19. 2022ULIP: Learning a Unified Representation of Language, Images, and Point Clouds for 3D Understandingcited by 267
  20. 2022Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-trainingcited by 264
  21. 2022Language-Grounded Indoor 3D Semantic Segmentation in the Wildcited by 246
  22. 2023Uni3D: Exploring Unified 3D Representation at Scalecited by 237
  23. 2021Self-Supervised Pretraining of 3D Features on any Point-Cloudcited by 217
  24. 2020Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contextscited by 210
  25. 2023ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understandingcited by 202
  26. 2020Unsupervised Point Cloud Pre-Training via Occlusion Completioncited by 186
  27. 2021Spatio-temporal Self-Supervised Representation Learning for 3D Point Cloudscited by 180
  28. 2023Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretrainingcited by 164
  29. 2022Masked Discrimination for Self-Supervised Learning on Point Cloudscited by 155
  30. 2022Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoderscited by 144
  31. 2024ShapeLLM: Universal 3D Object Understanding for Embodied Interactioncited by 137
  32. 2023Meta-Transformer: A Unified Framework for Multimodal Learningcited by 136
  33. 2021Uni-Perceiver: Pre-training Unified Architecture for Generic Perception for Zero-shot and Few-shot Taskscited by 125
  34. 2022Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?cited by 122
  35. 2023PointGPT: Auto-regressively Generative Pre-training from Point Cloudscited by 115
  36. 2023Swin3D: A Pretrained Transformer Backbone for 3D Indoor Scene Understandingcited by 113
  37. 2023OctFormer: Octree-based Transformers for 3D Point Cloudscited by 109
  38. 2023OneFormer3D: One Transformer for Unified Point Cloud Segmentationcited by 95

50 most cited of 74 All 74 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 3.

6 papers.

  1. 2025Chorus: Multi-Teacher Pretraining for Holistic 3D Gaussian Scene Encodingcited by 10
  2. 2025Investigating Domain Gaps for Indoor 3D Object Detectioncited by 2
  3. 2026Rethinking Point Clouds as Sequences: A Causal Next-Token Predictive Learning Frameworkcited by 1
  4. 2025Dual-Branch Center-Surrounding Contrast: Rethinking Contrastive Learning for 3D Point Cloudscited by 0
  5. 2026G2P: Gaussian-to-Point Attribute Alignment for Boundary-Aware 3D Segmentationcited by 0
  6. 2026Deformation-based In-Context Learning for Point Cloud Understandingcited by 0