DECA icon indicating copy to clipboard operation
DECA copied to clipboard

DECA: Detailed Expression Capture and Animation (SIGGRAPH 2021)

Results 168 DECA issues
Sort by recently updated
recently updated
newest added

Thank you for your contribution to the community. Can I ask you a question: what is the order of the expression blendshape? thanks

hello, thanks for sharing i want to use your model to predict AU intensity [(FACS)](https://en.wikipedia.org/wiki/Facial_Action_Coding_System) , is it possible ? could you give me some suggestion ?

Hi @YadiraF I am wondering what does this function do? I have an error in this line: mask[vis_parsing_anno>0.5] = 1. mask[vis_parsing_anno>0.5] = 1. TypeError: '>' not supported between instances of...

Hi, in your paper, you use a data cleaning strategy to filter some training images. Can you share this part of code? Thank you!

Thanks for your code. When I train the detail process, the loss of mrf keep around 12 from the beginning to the end, is it normal ?

I want to know is it normal for mrf to stay around 12? And I found that the verification phase of the detailed model is not as good as the...

I want to know how to get this file: datafile ='/ ps/scratch/face2d3d/texture_in_the_wild_code/VGGFace2_cleaning_codes/ringnetpp_training_lists/second_cleaning/vggface2_train_list_max_normal_100_ring_5_1_serial.npy' And How do I get other npy files?

As I thought generating full FLAME result is one of the strong points of DECA, I added some code to show the full FLAME image as a result. @YadiraF ![image](https://user-images.githubusercontent.com/79582852/137831706-0ec392cf-a6dd-4883-9b18-5b9a13af8bfb.png)...

Hi there, i run the tutorial from this page, ![image](https://user-images.githubusercontent.com/17308097/137735870-7e2ee71f-d447-406a-b262-95590c0c97a5.png) but it seem failed when it build... could you help me to fix that? Thanks ![image](https://user-images.githubusercontent.com/17308097/137738737-ca9d0236-da9d-481d-a60d-a56947998caa.png)

What should I do when I am in such a situation when I a ![000000](https://user-images.githubusercontent.com/80998075/136529094-ddaf01dd-1159-4ec3-b77e-c9cd51043cc0.jpg) ![001000](https://user-images.githubusercontent.com/80998075/136529107-6e1d9dac-8334-43cb-a33d-e48237cd900a.jpg) m training? And it happens all the time, and I need help.