基于 GSConv 的 Slim-neck：一种面向实时检测器架构的轻量化设计
Slim-neck by GSConv: A lightweight-design for real-time detector architectures
AI总结：
针对边缘设备实时检测中轻量化与精度难以兼顾的问题，提出 GSConv 卷积及 Slim-Neck 颈部设计，在二十余组实验中以约100FPS达到SODA10M 70.9% AP50。
摘要：
实时目标检测对于工业和研究领域具有重要意义。在边缘设备上，庞大的模型难以满足实时检测需求，而由大量深度可分离卷积构建的轻量级模型又无法达到足够的精度。我们提出一种新的轻量化卷积技术 GSConv，以在减轻模型负担的同时保持精度。GSConv 在精度和速度之间实现了出色的权衡。此外，我们还提供了一种基于 GSConv 的设计建议 Slim-Neck（SNs），以实现实时检测器更高的计算成本效益。SNs 的有效性在二十多组对比实验中得到了稳健验证。特别是，经 SNs 改进的实时检测器与基线相比达到了当前最优水平（在 Tesla T4 上以约 100FPS 的速度在 SODA10M 上取得 70.9% AP50）。代码可在 https://github.com/alanli1997/slim-neck-by-gsconv 获取。
英文摘要：
Real-time object detection is significant for industrial and research fields. On edge devices, a giant model is difficult to achieve the real-time detecting requirement and a lightweight model built from a large number of the depth-wise separable convolutional could not achieve the sufficient accuracy. We introduce a new lightweight convolutional technique, GSConv, to lighten the model but maintain the accuracy. The GSConv accomplishes an excellent trade-off between the accuracy and speed. Furthermore, we provide a design suggestion based on the GSConv, Slim-Neck (SNs), to achieve a higher computational cost-effectiveness of the real-time detectors. The effectiveness of the SNs was robustly demonstrated in over twenty sets comparative experiments. In particular, the real-time detectors of ameliorated by the SNs obtain the state-of-the-art (70.9% AP50 for the SODA10M at a speed of ~ 100FPS on a Tesla T4) compared with the baselines. Code is available at https://github.com/alanli1997/slim-neck-by-gsconv
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
