xiaoxiongli

during inference period, why you use SoftmaxActivation in P_Net and SoftmaxOutput in R_Net and ONet?

Dear @Seanlinx i feel confuse that why your demo.py do not draw 5 landmark, is your model trained to just detect face? Can i use your train code to train...

when i train my data, the category count is not 80 as coco, which code should i modify?

12

Dear all: when i train my data, the category count is not 80 as coco, which code should i modify? i find three part of code using "80" : 1)...

How to build in windows?

2

Hi Dear @rys : it seems can use VS2019 to build this project, but I do not find the .sln file, it is only have a CMakeList.txt, so Could you...

Why the highest training accuracy happened in Iteration 0 ?

7

Now I am training the 4th R-FCN stage (totally 4 stage, with OHEM, 229 category LOGOs), and I met a situation that the highest training accuracy(0.949) happened in Iteration 0...

Could this LUT model run on GPU?

1

Hi @yhjo09 I wanna to know that this LUT model(.npy) is run on CPU or GPU? if it is run on CPU, it is possiable to run on GPU for...

Why the PSNR value decreases by several DBs when I introduces H.264 encoding losses?

1

In my application scenario, the RGB data captured by the camera is down-scale(x2) at the transmit phase, then the down-sampled(x2) image is encoded by H.264 and transmitted over the network,...

why is so slow? in one V100 GPU, 10 second per frame?

2

Hi, @all, this is really good work, and I test it on DIV2k image, and 4 times SR from about resolution 400x500 --> 1600x2000, but is seems really slow: in...

it is easy to save memory (reduce to 10%) in function do_proprosal_test, but i find that the function fast_rcnn_train(mainly in function fast_rcnn_prepare_image_roidb) also cost a lot of cpu memory (about...