Donglearner

Dear scholar, Could you provide your 10 images from Google Image or the image weights? 

Dear scholar,   I don't know in your train36.h5df the semantic_matrix 's number reffered to which relation type. 0: wearing, 1: holding, 2: sitting on, 3: standing on, 4:...

Some bug in eval.py on Raw images The parameter didn't match your code eval.py......

Some wrong description in https://github.com/microsoft/Oscar/blob/master/MODEL_ZOO.md

1


Dear scholar, I only find the vqa and gqa image features in Oscar. But I can't find the download page about vqa and gqa image features from VinVL new detector.

Dear scholar, I want to ask whether your elegant code includes the function about produce a description about the attribute and object for the designated bounding box. In your tools/demo_image.py...

Dear scholar, I have doubts about one mathematical symbol in your classical paper "bi-att-flow". In the paper ,you said the dimension of similar matrix S is T times J. But...

Another download link? I can't download vcr task "train_boxes36.h5"(22.4G)

1


 

It doesn't realize two layers Lstm. Because the code only shows the bias = 2. It doesn't work. And The code only realize one layer Lstm or gru.