# MVS-TTA: Test-Time Adaptation for Multi-View Stereo via Meta-Auxiliary Learning

arXiv: https://arxiv.org/abs/2511.18120 · HTML: https://mlsanity.com/paper/2511.18120/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 35 references that are in the map.
- First posted on arXiv 22 Nov 2025.

## Builds on (most cited first, 26 of 26)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks](https://mlsanity.com/paper/1703.03400/) (2017)
- [VGGT: Visual Geometry Grounded Transformer](https://mlsanity.com/paper/2503.11651/) (2025)
- [MVSNet: Depth Inference for Unstructured Multi-view Stereo](https://mlsanity.com/paper/1804.02505/) (2018)
- [Twins: Revisiting the Design of Spatial Attention in Vision Transformers](https://mlsanity.com/paper/2104.13840/) (2021)
- [Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching](https://mlsanity.com/paper/1912.06378/) (2019)
- [A Comprehensive Survey on Test-Time Adaptation under Distribution Shifts](https://mlsanity.com/paper/2303.15361/) (2023)
- [PatchmatchNet: Learned Multi-View Patchmatch Stereo](https://mlsanity.com/paper/2012.01411/) (2020)
- [TransMVSNet: Global Context-aware Multi-view Stereo Network with Transformers](https://mlsanity.com/paper/2111.14600/) (2021)
- [A Survey on Deep Learning Techniques for Stereo-based Depth Estimation](https://mlsanity.com/paper/2006.02535/) (2020)
- [MVSFormer: Multi-View Stereo by Learning Robust Image Features and Temperature-based Depth](https://mlsanity.com/paper/2208.02541/) (2022)
- [MVSFormer++: Revealing the Devil in Transformer's Details for Multi-View Stereo](https://mlsanity.com/paper/2401.11673/) (2024)
- [An Alternative Probabilistic Interpretation of the Huber Loss](https://mlsanity.com/paper/1911.02088/) (2019)
- [Learning to Adapt for Stereo](https://mlsanity.com/paper/1904.02957/) (2019)
- [MetaGCD: Learning to Continually Learn in Generalized Category Discovery](https://mlsanity.com/paper/2308.11063/) (2023)
- [Learning Unsupervised Multi-View Stereopsis via Robust Photometric Consistency](https://mlsanity.com/paper/1905.02706/) (2019)
- [Adaptive Fusion of Single-View and Multi-View Depth for Autonomous Driving](https://mlsanity.com/paper/2403.07535/) (2024)
- [Self-supervised Learning of Depth Inference for Multi-view Stereo](https://mlsanity.com/paper/2104.02972/) (2021)
- [MVSAnywhere: Zero-Shot Multi-View Stereo](https://mlsanity.com/paper/2503.22430/) (2025)
- [Point-TTA: Test-Time Adaptation for Point Cloud Registration Using Multitask Meta-Auxiliary Learning](https://mlsanity.com/paper/2308.16481/) (2023)
- [Learning-based Multi-View Stereo: A Survey](https://mlsanity.com/paper/2408.15235/) (2024)
- [Learning to Adapt Frozen CLIP for Few-Shot Test-Time Domain Adaptation](https://mlsanity.com/paper/2506.17307/) (2025)
- [SDL-MVS: View Space and Depth Deformable Learning Paradigm for Multi-View Stereo Reconstruction in Remote Sensing](https://mlsanity.com/paper/2405.17140/) (2024)

## Abstract (arXiv)

Recent learning-based multi-view stereo (MVS) methods are data-driven and have achieved remarkable progress due to large-scale training data and advanced architectures. However, their generalization remains sub-optimal due to fixed model parameters trained on limited training data distributions. In contrast, optimization-based methods enable scene-specific adaptation but lack scalability and require costly per-scene optimization. In this paper, we propose MVS-TTA, an efficient test-time adaptation (TTA) framework that enhances the adaptability of learning-based MVS methods by bridging these two paradigms. Specifically, MVS-TTA employs a self-supervised, cross-view consistency loss as an auxiliary task to guide inference-time adaptation. We introduce a meta-auxiliary learning strategy to train the model to benefit from auxiliary-task-based updates explicitly. Our framework is model-agnostic and can be applied to a wide range of MVS methods with minimal architectural changes. Extensive experiments on standard datasets (DTU, BlendedMVS) and a challenging cross-dataset generalization setting demonstrate that MVS-TTA consistently improves performance, even when applied to state-of-the-art MVS models. To our knowledge, this is the first attempt to integrate optimization-based test-time adaptation into learning-based MVS using meta-learning. The code will be available at https://github.com/mart87987-svg/MVS-TTA.