# Breaking the Model Forgetting Cycle in Long-Incremental 3D Object Detection

arXiv: https://arxiv.org/abs/2607.14560 · HTML: https://mlsanity.com/paper/2607.14560/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 36 of its 45 references that are in the map.
- First posted on arXiv 16 Jul 2026.

## Builds on (most cited first, 30 of 36)

- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [Learning to Prompt for Continual Learning](https://mlsanity.com/paper/2112.08654/) (2021)
- [Deep Hough Voting for 3D Object Detection in Point Clouds](https://mlsanity.com/paper/1904.09664/) (2019)
- [CODA-Prompt: COntinual Decomposed Attention-based Prompting for Rehearsal-Free Continual Learning](https://mlsanity.com/paper/2211.13218/) (2022)
- [An End-to-End Transformer Model for 3D Object Detection](https://mlsanity.com/paper/2109.08141/) (2021)
- [Group-Free 3D Object Detection via Transformers](https://mlsanity.com/paper/2104.00678/) (2021)
- [Class-Incremental Learning: A Survey](https://mlsanity.com/paper/2302.03648/) (2023)
- [Incremental Learning of Object Detectors without Catastrophic Forgetting](https://mlsanity.com/paper/1708.06977/) (2017)
- [FCAF3D: Fully Convolutional Anchor-Free 3D Object Detection](https://mlsanity.com/paper/2112.00322/) (2021)
- [SESS: Self-Ensembling Semi-Supervised 3D Object Detection](https://mlsanity.com/paper/1912.11803/) (2019)
- [Overcoming Catastrophic Forgetting in Incremental Object Detection via Elastic Response Distillation](https://mlsanity.com/paper/2204.02136/) (2022)
- [SD-LoRA: Scalable Decoupled Low-Rank Adaptation for Class Incremental Learning](https://mlsanity.com/paper/2501.13198/) (2025)
- [Faster ILOD: Incremental Learning for Object Detectors based on Faster RCNN](https://mlsanity.com/paper/2003.03901/) (2020)
- [Continual Detection Transformer for Incremental Object Detection](https://mlsanity.com/paper/2304.03110/) (2023)
- [CAGroup3D: Class-Aware Grouping for 3D Object Detection on Point Clouds](https://mlsanity.com/paper/2210.04264/) (2022)
- [Cubify Anything: Scaling Indoor 3D Object Detection](https://mlsanity.com/paper/2412.04458/) (2024)
- [V-DETR: DETR with Vertex Relative Position Encoding for 3D Object Detection](https://mlsanity.com/paper/2308.04409/) (2023)
- [TR3D: Towards Real-Time Indoor 3D Object Detection](https://mlsanity.com/paper/2302.02858/) (2023)
- [Augmented Box Replay: Overcoming Foreground Shift for Incremental Object Detection](https://mlsanity.com/paper/2307.12427/) (2023)
- [UniDet3D: Multi-dataset Indoor 3D Object Detection](https://mlsanity.com/paper/2409.04234/) (2024)
- [Static-Dynamic Co-Teaching for Class-Incremental 3D Object Detection](https://mlsanity.com/paper/2112.07241/) (2021)
- [Unlocking Textual and Visual Wisdom: Open-Vocabulary 3D Object Detection Enhanced by Comprehensive Guidance from Text and Image](https://mlsanity.com/paper/2407.05256/) (2024)
- [Learning Class Prototypes for Unified Sparse Supervised 3D Object Detection](https://mlsanity.com/paper/2503.21099/) (2025)
- [Dual-Perspective Knowledge Enrichment for Semi-Supervised 3D Object Detection](https://mlsanity.com/paper/2401.05011/) (2024)
- [Continuous Subspace Optimization for Continual Learning](https://mlsanity.com/paper/2505.11816/) (2025)
- [VGGT-Det: Mining VGGT Internal Priors for Sensor-Geometry-Free Multi-View Indoor 3D Object Detection](https://mlsanity.com/paper/2603.00912/) (2026)
- [CT3D++: Improving 3D Object Detection with Keypoint-induced Channel-wise Transformer](https://mlsanity.com/paper/2406.08152/) (2024)

## Abstract (arXiv)

Incremental 3D object detection requires a detector to learn novel object classes while remembering previously learned ones over sequentially arriving data. Previous methods, primarily based on pseudo-labeling, perform reasonably in short-incremental stages but still suffer from severe model forgetting when dealing with long-incremental sequences. We investigate this failure and reveal a detrimental self-reinforcing cycle: data distribution shift of novel classes causes model forgetting on old classes, which further produces accumulated error in pseudo-labeling that exacerbates model degradation. To address this issue, we draw inspiration from the human learning process and propose the Learning-Dynamics-driven Memory and Review (LDMR) framework. LDMR monitors per-class detection quality at periodic training checkpoints and uses these learning-dynamics signals to drive two innovative mechanisms, namely (i) human-like intra-stage review that divides each incremental stage into multiple sub-stages' training and concentrates on remembering the most-forgotten objects, and (ii) scene-aware cross-stage memory evolution that evolves a memory bank to transfer knowledge between two consecutive stages by jointly considering scene learnability and diversity. Extensive experiments across multiple long-incremental protocols on indoor benchmarks SUN RGB-D and ScanNetV2 show that LDMR substantially mitigates the model forgetting and outperforms all baselines by a clear margin. Code is available at https://github.com/qianpeisheng/LDMR.