mlsanityData to 7 Oct 2026

CP3: Unifying Point Cloud Completion by Pretrain-Prompt-Predict Paradigm

arXiv 2207.05359 ↗First posted 12 Jul 2022cs.CVcited by 9

Not on the map. Why?

Abstract (arXiv)

Point cloud completion aims to predict complete shape from its partial observation. Current approaches mainly consist of generation and refinement stages in a coarse-to-fine style. However, the generation stage often lacks robustness to tackle different incomplete variations, while the refinement stage blindly recovers point clouds without the semantic awareness. To tackle these challenges, we unify point cloud Completion by a generic Pretrain-Prompt-Predict paradigm, namely CP3. Inspired by prompting approaches from NLP, we creatively reinterpret point cloud generation and refinement as the prompting and predicting stages, respectively. Then, we introduce a concise self-supervised pretraining stage before prompting. It can effectively increase robustness of point cloud generation, by an Incompletion-Of-Incompletion (IOI) pretext task. Moreover, we develop a novel Semantic Conditional Refinement (SCR) network at the predicting stage. It can discriminatively modulate multi-scale refinement with the guidance of semantics. Finally, extensive experiments demonstrate that our CP3 outperforms the state-of-the-art methods with a large margin.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 38 of its 53 references that are in the map. First posted on arXiv 12 Jul 2022.

Builds on

38 of 53 references.

References · 38

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  4. 2022Generative Adversarial Networkscited by 5,603
  5. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  6. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  7. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  8. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,769
  9. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,747
  10. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
  11. 2021Learning to Prompt for Vision-Language Modelscited by 2,269
  12. 2021Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processingcited by 2,222
26 more
  1. 2014Conditional Generative Adversarial Netscited by 2,152
  2. 2017Arbitrary Style Transfer in Real-time with Adaptive Instance Normalizationcited by 1,998
  3. 2019Language Models as Knowledge Bases?cited by 1,922
  4. 2021Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivitycited by 1,122
  5. 2016Generative Adversarial Text to Image Synthesiscited by 735
  6. 2018AtlasNet: A Papier-Mach'e Approach to Learning 3D Surface Generationcited by 539
  7. 2022Masked Autoencoders for Point Cloud Self-supervised Learningcited by 456
  8. 2018PCN: Point Completion Networkcited by 382
  9. 2018Recovering Realistic Texture in Image Super-resolution by Deep Spatial Feature Transformcited by 369
  10. 2021PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformerscited by 284
  11. 2020GRNet: Gridding Residual Network for Dense Point Cloud Completioncited by 174
  12. 2021SnowflakeNet: Point Cloud Completion by Snowflake Point Deconvolution with Skip-Transformercited by 162
  13. 2019Morphing and Sampling Network for Dense Point Cloud Completioncited by 132
  14. 2020Cascaded Refinement Network for Point Cloud Completioncited by 112
  15. 2021Density-aware Chamfer Distance as a Comprehensive Metric for Point Cloud Completioncited by 111
  16. 2021Variational Relational Point Completion Networkcited by 103
  17. 2021A Conditional Point Diffusion-Refinement Paradigm for 3D Point Cloud Completioncited by 100
  18. 2020PMP-Net: Point Cloud Completion by Learning Multi-step Point Moving Pathscited by 97
  19. 2019SG-NN: Sparse Generative Neural Networks for Self-Supervised Scene Completion of RGB-D Scanscited by 77
  20. 2020Conditional Sequential Modulation for Efficient Global Image Retouchingcited by 71
  21. 2020Detail Preserved Point Cloud Completion via Separated Feature Aggregationcited by 70
  22. 2020SoftPoolNet: Shape Descriptor for Point Cloud Completion and Classificationcited by 33
  23. 2020Cascaded Refinement Network for Point Cloud Completion with Self-supervisioncited by 20
  24. 2022Learning Local Displacements for Point Cloud Completioncited by 16
  25. 2021Very Lightweight Photo Retouching Network with Conditional Sequential Modulationcited by 14
  26. 2018Local Frequency Interpretation and Non-Local Self-Similarity on Graph for Point Cloud Inpaintingcited by 13

Built on by

Cited by year

Citing papers per year: 2023: 3; 2024: 2; 2025: 4.

9 papers.

  1. 2025PointSea: Point Cloud Completion via Self-structure Augmentationcited by 9
  2. 20243D Shape Completion on Unseen Categories:A Weakly-supervised Approachcited by 6
  3. 2023Fine-grained Text and Image Guided Point Cloud Completion with CLIP Modelcited by 5
  4. 2024Deep Learning for 3D Point Cloud Enhancement: A Surveycited by 4
  5. 2023Variational Relational Point Completion Network for Robust 3D Classificationcited by 3
  6. 2023GLiDR: Topologically Regularized Graph Generative Network for Sparse LiDAR Point Cloudscited by 2
  7. 2025SLACK: Attacking LiDAR-based SLAM with Adversarial Point Injectionscited by 1
  8. 2025Preserving Topological and Geometric Embeddings for Point Cloud Recoverycited by 0
  9. 2025CasPoinTr: Point Cloud Completion with Cascaded Networks and Knowledge Distillationcited by 0