search

luo3300612 / image-captioning-DLCT

Official pytorch implementation of paper "Dual-Level Collaborative Transformer for Image Captioning" (AAAI 2021).
BSD 3-Clause "New" or "Revised" License
194 stars 31 forks source link

how are alignment graph obtained for new datasets #19

Closed Davidwdq closed 2 years ago

Davidwdq commented 2 years ago

Hi, in your coding,h5py.File features has keys like ['%d_features' % image_id] , ['%d_grids' % image_id], ['%d_boxes' % image_id], ['%d_size' % image_id], ['%d_mask' % image_id], If I have a new data set, can I just use align.py to get geometric alignment graph after I get grid features and region features using extract_region_feature.py and grid-feats-vqa.

luo3300612 commented 2 years ago

Yes, all codes for generating geometric alignment graph are in align.ipynb with visualization function.