splendid

Why I can't get any match correspondence even with the same images?

10

Thank you for sharing this implementaion! ### Environment * **Platform**: Windows 10 64-bit, VS2015 OpenCV3.4.1 CUDA8 1060 ### Current Behavior: I have successfully build this repository with no errors. I...

Proactively let negative samples participate in training problems(主动让负样本参与训练问题)

3

Hi guys, In engineering projects, there are often situations in which a picture is completely unmarked. In the process of detecting, there are many misidentified results. Previously, in the darknet...

[Training issue]为什么在某些分辨率下，算法会对一类目标漏检？

7

up主你好，感谢你的工作。 1.我最近在用一些自己的数据训练Yolov3_lite，然后进行测试。因为数据图像的尺度（1280*1280）固定，目标尺寸大小固定，所以我重新更改了网络结构，并用 > It is very strange , can you try this [way](https://github.com/AlexeyAB/darknet/issues/1517) ? 的办法重新计算了anchor boxes，但是检测结果不理想，在测试集中该类目标一个都没检测出来！其他目标检测结果都正常； 2.此外，我还试过将图像分辨率（1024*1024）改小，并重新计算anchor，结果在该分辨率下，目标能检测出来，但是置信度较低。 3.若是将图像分辨率改为832*832，则该类目标检测结果正常。 My train prototxt name: "MobileNet-YOLO" layer { name: "data" type: "AnnotatedData" top:...

I have trained this model in ic15 with code in this repository. After 69100 iteration training, the loss is reduced to 0.00333686. >I1024 22:11:46.818552 4375 sgd_solver.cpp:105] Iteration 69180, lr =...

最终层的输出结果是NAN

5

同学你好，感谢你开源EAST-caffe，很漂亮的工作。 但是我在调试的时候，用caffe作为框架，发现最终层（f_score和geo_concat）的值全部都是NAN。于是我逐层输出每一层的feature map，发现conv_11_pw_hswish开始的输出中包含INF，并且之后的层里INF越来越多，导致最终无法得到想要的结果。 我的系统是ubuntu16，caffe是weiliu89/SSD的caffe，opencv版本4.1.1，RELU6层直接用普通的RELU代替了。 请问下你的RELU6层自己重新实现的，还是也用的普通RELU代替？