[英]False positives in faster-rcnn object detection
I'm training an object detector using tensorflow and the faster_rcnn_inception_v2_coco
model and am experiencing a lot of false positives when classifying on a video. 我正在使用tensorflow和
faster_rcnn_inception_v2_coco
模型训练一个物体探测器,并且在对视频进行分类时遇到很多误报。
After some research I've figured out that I need to add negative images to the training process. 经过一些研究,我发现我需要在训练过程中添加负面图像。
How do I add these to tfrecord
files? 如何将这些添加到
tfrecord
文件? I used the csv to tfrecord
file code provided in the tutorial here . 我在这里使用了教程中提供的csv到
tfrecord
文件代码。
Also it seems that ssd has a hard_example_miner
in the config that allows to configure this behaviour but this doesn't seem to be the case for faster rcnn? 此外,似乎ssd在配置中有一个
hard_example_miner
允许配置此行为,但这似乎不是更快的rcnn的情况? Is there a way to achieve something similar on faster rcnn? 有没有办法在更快的rcnn上实现类似的东西?
I was facing the same issue with faster RCNN, although you cannot actually use hard_example_miner with the faster RCNN model, you can add some background images , ie. 我用更快的RCNN遇到了同样的问题,虽然你实际上不能 将hard_example_miner与更快的RCNN模型一起使用,但你可以添加一些背景图像 ,即。 images with no objects (Everything remains the same, except there is not object tag in the xml for that particular picture)
没有对象的图像(一切都保持不变,除了xml中没有该特定图片的对象标记)
One more thing that actually worked wonders for me was using the imgaug library , you can augment the images and the bounding boxes using the same script . 实际上对我来说奇迹的另一件事是使用imgaug库 ,你可以使用相同的脚本扩充图像和边界框 。 Try and increase the training data by 10 or 15 times, and then I would suggest you to train again to around 150000-200000 steps.
尝试将训练数据增加10或15倍,然后我建议你再次训练大约150000-200000步。
These two steps helped me reduce the number of false positives effectively. 这两个步骤帮助我有效地减少了误报的数量。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.