Image Matching Across Wide Baselines: From Paper to Practice
Image Matching Across Wide Baselines: From Paper to Practice
复制标题
DOI：
10.1007/s11263-020-01385-0
复制
发表时间：
2020-10-07
影响因子：
19.5
通讯作者：
Trulls, Eduard
中科院分区：
文献类型：
--
作者：
Jin, Yuhe;Mishkin, Dmytro;Trulls, Eduard
我们引入了一个全面的基准本地功能和强大的估计算法，专注于下游的任务重建相机姿态的准确性作为我们的主要指标。我们的管道模块化结构允许轻松集成、配置和组合不同的方法和工艺。这是通过嵌入数十种流行的算法并对其进行评估来证明的，从开创性的作品到机器学习研究的前沿。我们表明，通过适当的设置，经典的解决方案可能仍然优于感知的最先进的状态。除了建立实际的最先进的状态，进行的实验揭示了意外的性能的结构从运动管道，可以帮助提高他们的性能，算法和学习的方法。数据和代码都是在线的（https：//github.com/ubcvision/image-matching-benchmark），提供了一个易于使用和灵活的框架，用于对本地特征和鲁棒的估计方法进行基准测试，无论是与顶级方法一起还是与顶级方法相比。这项工作为图像匹配挑战赛（https：//www.example.com）提供了基础。image-matching-challenge.github.io
We introduce a comprehensive benchmark for local features and robust estimation algorithms, focusing on the downstream task-the accuracy of the reconstructed camera pose-as our primary metric. Our pipeline's modular structure allows easy integration, configuration, and combination of different methods and heuristics. This is demonstrated by embedding dozens of popular algorithms and evaluating them, from seminal works to the cutting edge of machine learning research. We show that with proper settings, classical solutions may still outperform the perceived state of the art. Besides establishing the actual state of the art, the conducted experiments reveal unexpected properties of structure from motion pipelines that can help improve their performance, for both algorithmic and learned methods. Data and code are online (https://github.com/ubcvision/image-matching-benchmark), providing an easy-to-use and flexible framework for the benchmarking of local features and robust estimation methods, both alongside and against top-performing methods. This work provides a basis for the Image Matching Challenge (https://image-matching-challenge.github.io).