# Fast-FoundationStereo: Real-Time Zero-Shot Stereo Matching

arXiv: https://arxiv.org/abs/2512.11130 · HTML: https://mlsanity.com/paper/2512.11130/

## Key facts

- Cited by 28 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 67 of its 90 references that are in the map.
- First posted on arXiv 11 Dec 2025.

## 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 67)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [VGGT: Visual Geometry Grounded Transformer](https://mlsanity.com/paper/2503.11651/) (2025)
- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [DINOv3](https://mlsanity.com/paper/2508.10104/) (2025)
- [Regularized Evolution for Image Classifier Architecture Search](https://mlsanity.com/paper/1802.01548/) (2018)
- [A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation](https://mlsanity.com/paper/1512.02134/) (2015)
- [DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset](https://mlsanity.com/paper/2403.12945/) (2024)
- [Importance Estimation for Neural Network Pruning](https://mlsanity.com/paper/1906.10771/) (2019)
- [Pyramid Stereo Matching Network](https://mlsanity.com/paper/1803.08669/) (2018)
- [Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching](https://mlsanity.com/paper/1912.06378/) (2019)
- [RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching](https://mlsanity.com/paper/2109.07547/) (2021)
- [Faster Segment Anything: Towards Lightweight SAM for Mobile Applications](https://mlsanity.com/paper/2306.14289/) (2023)
- [Unifying Flow, Stereo and Depth Estimation](https://mlsanity.com/paper/2211.05783/) (2022)
- [Group-wise Correlation Stereo Network](https://mlsanity.com/paper/1903.04025/) (2019)
- [DepGraph: Towards Any Structural Pruning](https://mlsanity.com/paper/2301.12900/) (2023)
- [Iterative Geometry Encoding Volume for Stereo Matching](https://mlsanity.com/paper/2303.06615/) (2023)
- [EfficientSAM: Leveraged Masked Image Pretraining for Efficient Segment Anything](https://mlsanity.com/paper/2312.00863/) (2023)
- [FoundationStereo: Zero-Shot Stereo Matching](https://mlsanity.com/paper/2501.09898/) (2025)
- [UniDepthV2: Universal Monocular Metric Depth Estimation Made Simpler](https://mlsanity.com/paper/2502.20110/) (2025)
- [Hierarchical Neural Architecture Search for Deep Stereo Matching](https://mlsanity.com/paper/2010.13501/) (2020)
- [Compact Language Models via Pruning and Knowledge Distillation](https://mlsanity.com/paper/2407.14679/) (2024)
- [Attention Concatenation Volume for Accurate and Efficient Stereo Matching](https://mlsanity.com/paper/2203.02146/) (2022)
- [Structured Pruning for Deep Convolutional Neural Networks: A survey](https://mlsanity.com/paper/2303.00566/) (2023)
- [EdgeNeXt: Efficiently Amalgamated CNN-Transformer Architecture for Mobile Vision Applications](https://mlsanity.com/paper/2206.10589/) (2022)
- [HITNet: Hierarchical Iterative Tile Refinement Network for Real-time Stereo Matching](https://mlsanity.com/paper/2007.12140/) (2020)

## Built on by (most cited first, 28 of 28)

- [LUCID: Learning Embodiment-Agnostic Intent Models from Unstructured Human Videos for Scalable Dexterous Robot Skill Acquisition](https://mlsanity.com/paper/2606.11628/) (2026)
- [VIA: Visual Interface Agent for Robot Control](https://mlsanity.com/paper/2607.11119/) (2026)
- [StereoPolicy: Improving Robotic Manipulation Policies via Stereo Perception](https://mlsanity.com/paper/2605.09989/) (2026)
- [HarnessVLN: Unifying Training-Free Embodied Navigation through an Agent Harness](https://mlsanity.com/paper/2609.15195/) (2026)
- [SCARED-C: Corrected Camera Poses for Endoscopic Depth Estimation](https://mlsanity.com/paper/2605.16628/) (2026)
- [See like a Robot: Robot-Centric Pointmaps for VLA Models](https://mlsanity.com/paper/2607.11498/) (2026)
- [ExStereo: Lifting 2D Vision-Language-Action Models to 3D with Explicit Stereo Representations](https://mlsanity.com/paper/2610.04805/) (2026)
- [OmniRobotHome: A Multi-Camera Home Platform for Real-Time Human-Robot Interaction](https://mlsanity.com/paper/2604.28197/) (2026)
- [H-OmniStereo: Zero-Shot Omnidirectional Stereo Matching with Heading-Aligned Normal Priors](https://mlsanity.com/paper/2605.14963/) (2026)
- [LadderMan: Learning Humanoid Perceptive Ladder Climbing](https://mlsanity.com/paper/2606.05873/) (2026)
- [Mana: Dexterous Manipulation of Articulated Tools](https://mlsanity.com/paper/2606.13677/) (2026)
- [StereoFactory: A Unified Merging Framework for Robust Stereo Matching](https://mlsanity.com/paper/2606.17475/) (2026)
- [Lite Any Stereo V2: Faster and Stronger Efficient Zero-Shot Stereo Matching](https://mlsanity.com/paper/2606.24457/) (2026)
- [SILO: Simulation-in-the-Loop Sim-to-Real Transfer for Multi-Stage Cable Routing](https://mlsanity.com/paper/2607.04616/) (2026)
- [URS-Stereo: Uncertainty-Guided Residual Search for Real-Time Stereo Matching](https://mlsanity.com/paper/2607.06779/) (2026)
- [WAVE-Stereo: Warp-Aligned Volume Encoding for Stereo Matching](https://mlsanity.com/paper/2607.13674/) (2026)
- [RayLift: Lifting Complementary Ray-Wise Evidence with 3D Geometry Priors for Semantic Scene Completion](https://mlsanity.com/paper/2608.08476/) (2026)
- [NBS: No Bias Stereo](https://mlsanity.com/paper/2608.28933/) (2026)
- [Physico-Geospatial Grounded Scene Interpretation for Mobile Robotics](https://mlsanity.com/paper/2609.06629/) (2026)
- [Visuomotor Robotic Pruning in Planar Orchards Using Hybrid Reinforcement Learning](https://mlsanity.com/paper/2609.24906/) (2026)
- [JAMB: Joint Action-Motion Diffusion for Bimanual Manipulation](https://mlsanity.com/paper/2609.25322/) (2026)
- [An Empirical Study on What Matters for Viewpoint-Generalizable Policies in Visual Imitation Learning](https://mlsanity.com/paper/2609.32762/) (2026)
- [Counterfactual Video Generation Enables Scalable Humanoid Loco-Manipulation](https://mlsanity.com/paper/2609.38172/) (2026)
- [StereoGaussians: Feed-Forward 3D Gaussian Splatting from Stereo Images](https://mlsanity.com/paper/2609.38592/) (2026)
- [PointWAM: 3D World Action Modeling for Dexterous Robotic Manipulation](https://mlsanity.com/paper/2610.02840/) (2026)
- [RLHND: Video Foundation Models as Physically Grounded Hand Trackers for Robot Learning](https://mlsanity.com/paper/2610.09455/) (2026)
- [Thinker-WM: A Unified World Model with Shared Multimodal Representations for Embodied Intelligence](https://mlsanity.com/paper/2610.18447/) (2026)
- [FoveaVLA: Foveation Actions Replace the Wrist Camera in VLAs](https://mlsanity.com/paper/2610.20064/) (2026)

## Abstract (arXiv)

Stereo foundation models achieve strong zero-shot generalization but remain computationally prohibitive for real-time applications. Efficient stereo architectures, on the other hand, sacrifice robustness for speed and require costly per-domain fine-tuning. To bridge this gap, we present Fast-FoundationStereo, a family of architectures that achieve, for the first time, strong zero-shot generalization at real-time frame rate. We employ a divide-and-conquer acceleration strategy with three components: (1) knowledge distillation to compress the hybrid backbone into a single efficient student; (2) blockwise neural architecture search for automatically discovering optimal cost filtering designs under latency budgets, reducing search complexity exponentially; and (3) structured pruning for eliminating redundancy in the iterative refinement module. Furthermore, we introduce an automatic pseudo-labeling pipeline used to curate 1.4M in-the-wild stereo pairs to supplement synthetic training data and facilitate knowledge distillation. The resulting model can run over 10x faster than FoundationStereo while closely matching its zero-shot accuracy, thus establishing a new state-of-the-art among real-time methods. Project page: https://nvlabs.github.io/Fast-FoundationStereo/