实时零样本立体匹配新模型，速度超10倍且精度接近顶尖水平。
Fast-FoundationStereo: Real-Time Zero-Shot Stereo Matching
- 分治加速：知识蒸馏+块级神经搜索+结构化剪枝
- 实测帧率超10倍于原模型，零样本精度近乎持平
- 自动生成140万野外立体图像对，提升泛化能力
立体基础模型虽具备强零样本泛化能力，但计算开销大，难以满足实时需求。高效立体架构则以牺牲鲁棒性为代价换取速度，且需高昂的领域微调成本。为此，我们提出 Fast-FoundationStereo，首次实现实时帧率下的强零样本泛化。采用分治加速策略：（1）知识蒸馏将混合骨干压缩为单一高效学生网络；（2）块级神经架构搜索在延迟预算下自动发现最优代价过滤设计，将搜索复杂度指数级降低；（3）结构化剪枝消除迭代优化模块中的冗余。此外，引入自动伪标注流水线，构建140万张野外立体图像对，补充合成数据并支持知识蒸馏。所提模型运行速度较 FoundationStereo 超10倍，同时零样本精度高度接近其表现，成为实时方法新基准。
原文摘要 · Abstract (English)
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/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。