COST
COST copied to clipboard
object lives extraction method
Thanks for the nice work! I am interested in the the object appearance and spatial feature extraction process. You briefly mentioned it in Sec. 4.1 on page 10. However, I didn't find the open-source implementation code of the reference, "Hierarchical Object-oriented Spatio-Temporal Reasoning for Video Question Answering". It do confuse me.
Could I ask you to do me a favor and provide the access to the object feature extraction code scripts or the obtained feature(line 24 mentioned in "preprocess/dump_h5.py")?
Look forward to your kind reply, and thanks a lot.