STTran icon indicating copy to clipboard operation
STTran copied to clipboard

Spatial-Temporal Transformer for Dynamic Scene Graph Generation, ICCV2021

Results 17 STTran issues
Sort by recently updated
recently updated
newest added

from fasterRCNN.lib.model import _C ImportError: /home/valca509/STTran-main/fasterRCNN/lib/model/_C.cpython-36m-x86_64-linux-gnu.so: undefined symbol: _ZN6caffe26detail36_typeMetaDataInstance_preallocated_7E

@yrcong hi, I would like to save the scene graph. Is it possible to save these graph like nodes and edges format? I don't know if the author realizes the...

I extract the format triplet prediction from the results, but how can I assign all the triplets to each frame?

你好,请问ObjectClassifier中的clean_class的class_idx为什么输入5,8,17呢

In order to test in a custom video, you said only video frames (self.video_list in the class AG) are necessary (for the setting SGDET), but can I get the result...

Hi, Thanks for your great work! I am curious how the visualization results in the paper are generated? Thanks

Hello, I would like to ask how your dataset is labeled, I want to label my own data, do you have any suggestions?