fangxu comments

Results 25 comments of


fangxu

Lack of SOTA temporal action detection models

@kennymckormick Hi, are you planning to add https://github.com/TencentYoutuResearch/ActionDetection-AFSD ?

Speed Scene translation and rotation error ?

but just path length plot, I don't find Speed (horizontal axis) figure.~~~

how plot different method predict result in one image ?

Yes， I mean ， If we have two or above method prediction result. Could this tool plot different method curves in one figure. like this ,it's not only ground truth...

请问，需要用matlab工具箱进行最极大抑制吗？

是之在进行预测之前需要使用 matlab 工具向作极大值抑制处理吗

multi_stage_meanfield.cpp:49] This implementation has not been tested batch size > 1.

The most critical thing is my batch_size=1., I use voc-layer load data。。。

multi_stage_meanfield.cpp:49] This implementation has not been tested batch size > 1.

Now, I use the ，The problem has gone away and a new problem has come.. ``` I0706 15:28:25.721647 31458 net.cpp:744] Ignoring source layer score_pool3 I0706 15:28:25.721652 31458 net.cpp:744] Ignoring source...

multi_stage_meanfield.cpp:49] This implementation has not been tested batch size > 1.

the fcn-model is the voc-fcn8s-atonce, I added the crfasRNN layer to fcn8s-atonce's train.prototxt , or the use TVG_CRFRNN_new_traintest.prototxt .it have the same problem...

multi_stage_meanfield.cpp:49] This implementation has not been tested batch size > 1.

不晓得能和你讲中文吗？

multi_stage_meanfield.cpp:49] This implementation has not been tested batch size > 1.

@bittnt

multi_stage_meanfield.cpp:49] This implementation has not been tested batch size > 1.

I think the net bottom.shape is the same .but i don't know what something wrong