Ayush Rai

Training on Different Dataset

10

@bgshih I am trying to train crnn on a synthetic dataset that I generated. Along with english alphabets and digits (0-9) it contains special characters like $,@,# etc. I was...

Preprocessing of Training Images

3

Hi @bgshih Is subtracting the mean and dividing by the standard deviation important preprocessing step in CRNN?

Using UTF-8 Character encoding instead of ASCII

1

@bgshih The text data that I want to recognize involves currency symbols like dollar sign, euro sign, Indian rupee sign etc. Most of these are not present in the ASCII...

Hi @KuangHaofei and @bryanyzhu I was looking at your paper. Could you please explain how to apply VCLR at video level and not on frame level? How do I modify...

Regarding CTPN's Performance on Multi-Oriented Text

1

@tianzhi0549 I figured out that the performance of CTPN on slanted/multi-oriented text is not very good. I want to tackle this by introducing the Spatial Transformer Networks just after the...

Poor Scene Graph Detection Results on Custom Images

16

## ❓ Questions and Help Hi @KaihuaTang Thanks for the code for scene graph detection. I was trying to run the code for scene graph generation on custom images but...

Extracting H.264 motion vectors

2

Hi guys, Thank you for such a cool library. Is there a way to extract H.264 motion vectors from a video using decord library? Any help would be really appreciated....

Hi @MhLiao I am successfully able to train TextBoxes on my custom dataset. But while training, I am not able to see the mbox loss in the validation (or test)...

Hello @xingyizhou Thank you for the work. I wanted to ask you if you calculate some sort of confidence regarding detection of the joints through confidence heat maps? How can...

Hello @xingyizhou, Thank you for a great 3d pose estimation model. Well I wanted to finetune the given pretrained model on other 2D pose and 3D pose datasets like NTU...