Qilin Li
Is there anyone can reproduce the performance scores, i can only get the segmentation image, but not the scores, due to a lot of code bugs in eval_acc.m.
in eval_acc.m, line 38, comment '' imAnno = imAnno + 1; " and try
What's your cuda and g++ version? Well, I see you only have GTX 970. If that is the case, you cannot run PSPNet test case due to the limit graphics...
sorry not the eval_all.m, should be eval_acc.m.
@BassyKuo I think I found the reason of error. It comes from the PSPNET-cudnn5/evaluation/evaluationCode/intersectionAndUnion. Since I am testing the pre-trained model on Pascal voc 2012 test set, there is no...
@MarcelSimon sorry didn;t make it clear. I mean in the prototxt cnn-models/ResNet_preact/ResNet50_cvgj/train.prototxt line 295-318, which is the first expanding layer. Yours are expanded from "layer_64_1_conv1". While in He's implementation deep-residual-networks/prototxt/ResNet-50-deploy.prototxt...
As you mentioned the sharing batch norm, scale, it reminds me another difference between yours and He's implementation. If you check ther implementation http://ethereon.github.io/netscope/#/gist/db945b393d40bfa26006 ( graph) or https://github.com/KaimingHe/deep-residual-networks/blob/master/prototxt/ResNet-50-deploy.prototxt (prototxt) they...
I see. Thanks a lot @MarcelSimon
is the problem solved?