mlsanityData to 14 Oct 2026

ReFu: Recursive Fusion for Exemplar-Free 3D Class-Incremental Learning

arXiv 2409.12326 ↗First posted 18 Sep 2024cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

We introduce a novel Recursive Fusion model, dubbed ReFu, designed to integrate point clouds and meshes for exemplar-free 3D Class-Incremental Learning, where the model learns new 3D classes while retaining knowledge of previously learned ones. Unlike existing methods that either rely on storing historical data to mitigate forgetting or focus on single data modalities, ReFu eliminates the need for exemplar storage while utilizing the complementary strengths of both point clouds and meshes. To achieve this, we introduce a recursive method which continuously accumulates knowledge by updating the regularized auto-correlation matrix. Furthermore, we propose a fusion module, featuring a Pointcloud-guided Mesh Attention Layer that learns correlations between the two modalities. This mechanism effectively integrates point cloud and mesh features, leading to more robust and stable continual learning. Experiments across various datasets demonstrate that our proposed framework outperforms existing methods in 3D class-incremental learning.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 52 of its 63 references that are in the map. First posted on arXiv 18 Sep 2024.

Builds on

52 of 63 references.

References · 52

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  4. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  5. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  6. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  7. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,897
  8. 2016Learning without Forgettingcited by 2,496
  9. 2016iCaRL: Incremental Classifier and Representation Learningcited by 2,333
  10. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,223
  11. 2018Experience Replay for Continual Learningcited by 1,095
  12. 2021Learning to Prompt for Continual Learningcited by 944
38 more
  1. 2023A Comprehensive Survey of Continual Learning: Theory, Method and Applicationcited by 872
  2. 2022DualPrompt: Complementary Prompting for Rehearsal-free Continual Learningcited by 642
  3. 2021PointCLIP: Point Cloud Understanding by CLIPcited by 463
  4. 2022Masked Autoencoders for Point Cloud Self-supervised Learningcited by 459
  5. 2020Class-incremental learning: survey and performance evaluation on image classificationcited by 449
  6. 2021Rainbow Memory: Continual Learning with a Memory of Diverse Samplescited by 273
  7. 2022FOSTER: Feature Boosting and Compression for Class-Incremental Learningcited by 229
  8. 2023Class-Incremental Learning: A Surveycited by 209
  9. 2023Revisiting Class-Incremental Learning with Pre-Trained Models: Generalizability and Adaptivity are All You Needcited by 184
  10. 2023RanPAC: Random Projections and Pre-trained Models for Continual Learningcited by 164
  11. 2024Continual Learning with Pre-Trained Models: A Surveycited by 154
  12. 2020Adaptive Aggregation Networks for Class-Incremental Learningcited by 149
  13. 2022Decorate the Newcomers: Visual Domain Prompt for Continual Test Time Adaptationcited by 98
  14. 2019Continual Learning for Robotics: Definition, Framework, Learning Strategies, Opportunities and Challengescited by 93
  15. 2021Subdivision-Based Mesh Convolution Networkscited by 60
  16. 2022ACIL: Analytic Class-Incremental Learning with Absolute Memorization and Privacy Protectioncited by 57
  17. 2022An Efficient Domain-Incremental Learning Approach to Drive in All Weather Conditionscited by 54
  18. 2024Consistent Prompting for Rehearsal-Free Continual Learningcited by 54
  19. 2023TAIL: Task-specific Adapters for Imitation Learning with Large Pretrained Modelscited by 53
  20. 2023Online Hyperparameter Optimization for Class-Incremental Learningcited by 36
  21. 2024Pre-trained Vision and Language Transformers Are Few-Shot Incremental Learnerscited by 36
  22. 2023Online Class Incremental Learning on Stochastic Blurry Task Boundary via Mask and Visual Prompt Tuningcited by 35
  23. 2023Continual Self-supervised Learning: Towards Universal Multi-modal Medical Data Representation Learningcited by 34
  24. 2023Audio-Visual Class-Incremental Learningcited by 33
  25. 2024Convolutional Prompting meets Language Models for Continual Learningcited by 32
  26. 2022MeshMAE: Masked Autoencoders for 3D Mesh Data Analysiscited by 31
  27. 2022Few-Shot Continual Active Learning by a Robotcited by 25
  28. 2020I3DOL: Incremental 3D Object Learning without Catastrophic Forgettingcited by 20
  29. 2022Do Pre-trained Models Benefit Equally in Continual Learning?cited by 20
  30. 2023Peer-to-Peer Federated Continual Learning for Naturalistic Driving Action Recognitioncited by 19
  31. 2023Adapt Your Teacher: Improving Knowledge Distillation for Exemplar-free Continual Learningcited by 16
  32. 2024Text-Enhanced Data-free Approach for Federated Class-Incremental Learningcited by 16
  33. 2022Few-shot Class-incremental Learning for 3D Point Cloud Objectscited by 13
  34. 2022CONDA: Continual Unsupervised Domain Adaptation Learning in Visual Perception for Self-Driving Carscited by 11
  35. 2023InOR-Net: Incremental 3D Object Recognition Network for Point Cloud Representationcited by 11
  36. 2024iNeMo: Incremental Neural Mesh Models for Robust Class-Incremental Learningcited by 10
  37. 2020Knowledge Capture and Replay for Continual Learningcited by 7
  38. 2021Learning without Forgetting for 3D Point Cloud Objectscited by 7

50 most cited of 52 All 52 (JSON)

Built on by

Not cited yet.