Lessons learned(经验教训)
Some key findings from the Google Research paper: 谷歌研究报告的一些重要发现:
- R-FCN and SSD models are faster on average but cannot beat the Faster R-CNN in accuracy if speed is not a concern.
- 一般而言,R-FCN和ssd的速度更快,但如果不考虑速度的话,其精度无法超过Faster R-CNN。
Faster R-CNN requires at least 100 ms per image. - Faster R-CNN每张图像至少要100毫秒。
- Use only low-resolution feature maps for detections hurts accuracy badly.
- 仅使用低分辨率特征图会严重降低准确度。
- Input image resolution impacts accuracy significantly. Reduce image size by half in width and height lowers accuracy by 15.88% on average but also reduces inference time by 27.4% on average.
- 输入图像的分辨率对准确度影响很大。将图像的宽度和高度减少一半,平均降低了15.88%的准确度,同时也平均减少了27.4%的推理时间。
- Choice of feature extractors impacts detection accuracy for Faster R-CNN and R-FCN but less reliant for SSD.
- 对于Faster R-CNN和R-FCN,特征提取器的选择会影响检测准确度,但SSD对此依赖性较小。
- Post processing includes non-max suppression (which only run on CPU) takes up the bulk of the running time for the fastest models at about 40 ms which caps speed to 25 FPS.
- 后期处理包括非极大值抑制(仅在CPU上运行)占最快模型推理时间的大部分,约为40毫秒,速度上限为25 FPS。
If mAP is calculated with one single IoU only, use mAP@IoU=0.75. - 如果mAP仅使用一个IoU计算,则使用mAP@IoU=0.75。
- With an Inception ResNet network as a feature extractor, the use of stride 8 instead of 16 improves the mAP by a factor of 5%, but increased running time by a factor of 63%.
- 使用Inception ResNet网络作为特征提取器,使用步长8而不是16将地图提高了5%,但将运行时间提高了63%。
Most accurate 最准确 - The most accurate single model use Faster R-CNN using Inception ResNet with 300 proposals. It runs at 1 second per image.
- 最精确的单一模型使用Faster R-CNN,使用Inception ResNet和300个候选框。它以每张图像1秒的速度运行。
- The most accurate model is an ensemble model with multi-crop inference. It achieves state-of-the-art detection on 2016 COCO challenge in accuracy. It uses the vector of average precision to select five most different models.
- 最精确的模型是具有多截断推断的集成模型。它在2016年COCO挑战赛上实现了最先进的检测。它使用平均精度向量来选择五种最不同的模型。
- Fastest 最快的
- SSD with MobileNet provides the best accuracy tradeoff within the fastest detectors.
- 带MobileNet的SSD在最快的检测器内提供最佳的精度权衡。
- SSD is fast but performs worse for small objects comparing with others.SSD速度很快,但对小对象的性能比其他对象差。
- For large objects, SSD can outperform Faster R-CNN and R-FCN in accuracy with lighter and faster extractors.
- 对于大型对象,SSD在精度上优于更快的R-CNN和R-FCN,具有更轻和更快的提取器。
Good balance between accuracy and speed 准确度和速度之间的良好平衡
- Faster R-CNN can match the speed of R-FCN and SSD at 32mAP if we reduce the number of proposal to 50.
-
如果我们将提案数量减少到50个,更快的R-CNN可以在32mAP上与R-FCN和SSD的速度相匹配。