论文标题
RF-Next：有效的接收现场搜索卷积神经网络
RF-Next: Efficient Receptive Field Search for Convolutional Neural Networks
论文作者
论文摘要
模型的时间/空间接受场在顺序/空间任务中起重要作用。大型接受场有助于长期关系，而小型接受场有助于捕获当地的细节。现有方法构建具有手工设计的接收场的模型。我们可以有效地搜索接收场合组合以取代手工设计的模式吗？为了回答这个问题，我们建议通过全球到本地搜索方案找到更好的接受场组合。我们的搜索方案利用了全局搜索以找到粗糙的组合和本地搜索，以进一步获得精致的接收场组合。全球搜索发现除了人类设计的模式以外的其他可能的粗糙组合。除了全球搜索之外，我们提出了一种预期的迭代局部搜索方案，以有效地完善组合。我们的RF-Next模型，将接受现场搜索插入各种模型，提高许多任务的性能，例如时间动作分割，对象检测，实例分割和语音综合。源代码在http://mmcheng.net/rfnext上公开可用。
Temporal/spatial receptive fields of models play an important role in sequential/spatial tasks. Large receptive fields facilitate long-term relations, while small receptive fields help to capture the local details. Existing methods construct models with hand-designed receptive fields in layers. Can we effectively search for receptive field combinations to replace hand-designed patterns? To answer this question, we propose to find better receptive field combinations through a global-to-local search scheme. Our search scheme exploits both global search to find the coarse combinations and local search to get the refined receptive field combinations further. The global search finds possible coarse combinations other than human-designed patterns. On top of the global search, we propose an expectation-guided iterative local search scheme to refine combinations effectively. Our RF-Next models, plugging receptive field search to various models, boost the performance on many tasks, e.g., temporal action segmentation, object detection, instance segmentation, and speech synthesis. The source code is publicly available on http://mmcheng.net/rfnext.
