# TTT-KD: Test-Time Training for 3D Semantic Segmentation through Knowledge Distillation from Foundation Models

arXiv: https://arxiv.org/abs/2403.11691 · HTML: https://mlsanity.com/paper/2403.11691/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Detection and dense prediction distillation, one of 12 directions in Knowledge distillation.
- Builds on 64 of its 76 references that are in the map.
- First posted on arXiv 18 Mar 2024.

## Placed in

- Knowledge distillation / [Detection and dense prediction distillation](https://mlsanity.com/distill/detection-and-dense-prediction-distillation/)

## Builds on (most cited first, 30 of 64)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://mlsanity.com/paper/1903.12261/) (2019)
- [Unsupervised Representation Learning by Predicting Image Rotations](https://mlsanity.com/paper/1803.07728/) (2018)
- [Point Transformer](https://mlsanity.com/paper/2011.00931/) (2020)
- [Matterport3D: Learning from RGB-D Data in Indoor Environments](https://mlsanity.com/paper/1709.06158/) (2017)
- [4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks](https://mlsanity.com/paper/1904.08755/) (2019)
- [KPConv: Flexible and Deformable Convolution for Point Clouds](https://mlsanity.com/paper/1904.08889/) (2019)
- [Vision Transformers Need Registers](https://mlsanity.com/paper/2309.16588/) (2023)
- [Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2002.08546/) (2020)
- [Point Transformer V3: Simpler, Faster, Stronger](https://mlsanity.com/paper/2312.10035/) (2023)
- [Continual Test-Time Domain Adaptation](https://mlsanity.com/paper/2203.13591/) (2022)
- [Efficient Test-Time Model Adaptation without Forgetting](https://mlsanity.com/paper/2204.02610/) (2022)
- [Masked Autoencoders for Point Cloud Self-supervised Learning](https://mlsanity.com/paper/2203.06604/) (2022)
- [Towards Stable Test-Time Adaptation in Dynamic Wild World](https://mlsanity.com/paper/2302.12400/) (2023)
- [Joint 2D-3D-Semantic Data for Indoor Scene Understanding](https://mlsanity.com/paper/1702.01105/) (2017)
- [Point Transformer V2: Grouped Vector Attention and Partition-based Pooling](https://mlsanity.com/paper/2210.05666/) (2022)
- [MEMO: Test Time Robustness via Adaptation and Augmentation](https://mlsanity.com/paper/2110.09506/) (2021)
- [Contrastive Test-Time Adaptation](https://mlsanity.com/paper/2204.10377/) (2022)
- [Test-Time Training with Masked Autoencoders](https://mlsanity.com/paper/2209.07522/) (2022)
- [Super-Convergence: Very Fast Training of Neural Networks Using Large Learning Rates](https://mlsanity.com/paper/1708.07120/) (2017)
- [A2D2: Audi Autonomous Driving Dataset](https://mlsanity.com/paper/2004.06320/) (2020)
- [Parameter-free Online Test-time Adaptation](https://mlsanity.com/paper/2201.05718/) (2022)
- [ST3D: Self-training for Unsupervised Domain Adaptation on 3D Object Detection](https://mlsanity.com/paper/2103.05346/) (2021)
- [Train in Germany, Test in The USA: Making 3D Object Detectors Generalize](https://mlsanity.com/paper/2005.08139/) (2020)

## Built on by (most cited first, 5 of 5)

- [Beyond Model Adaptation at Test Time: A Survey](https://mlsanity.com/paper/2411.03687/) (2024)
- [Improving Multimodal Distillation for 3D Semantic Segmentation under Domain Shift](https://mlsanity.com/paper/2511.17455/) (2025)
- [Hierarchical Adaptive networks with Task vectors for Test-Time Adaptation](https://mlsanity.com/paper/2508.09223/) (2025)
- [Hierarchical Mesh Transformers with Topology-Guided Pretraining for Morphometric Analysis of Brain Structures](https://mlsanity.com/paper/2604.05215/) (2026)
- [COSTA: A Cluster-Centric Paradigm for Annotation-Free Open-Set Semantic Segmentation of Aerial Point Clouds with Domain Shifts](https://mlsanity.com/paper/2608.18479/) (2026)

## Abstract (arXiv)

Test-Time Training (TTT) proposes to adapt a pre-trained network to changing data distributions on-the-fly. In this work, we propose the first TTT method for 3D semantic segmentation, TTT-KD, which models Knowledge Distillation (KD) from foundation models (e.g. DINOv2) as a self-supervised objective for adaptation to distribution shifts at test-time. Given access to paired image-pointcloud (2D-3D) data, we first optimize a 3D segmentation backbone for the main task of semantic segmentation using the pointclouds and the task of 2D → 3D KD by using an off-the-shelf 2D pre-trained foundation model. At test-time, our TTT-KD updates the 3D segmentation backbone for each test sample, by using the self-supervised task of knowledge distillation, before performing the final prediction. Extensive evaluations on multiple indoor and outdoor 3D segmentation benchmarks show the utility of TTT-KD, as it improves performance for both in-distribution (ID) and out-of-distribution (ODO) test datasets. We achieve a gain of up to 13% mIoU (7% on average) when the train and test distributions are similar and up to 45% (20% on average) when adapting to OOD test samples.