search

fredzzhang / upt

[CVPR'22] Official PyTorch implementation for paper "Efficient Two-Stage Detection of Human–Object Interactions with a Novel Unary–Pairwise Transformer"
https://fredzzhang.com/unary-pairwise-transformers
BSD 3-Clause "New" or "Revised" License
144 stars 26 forks source link

VCOCO evaluation #86

Closed aqingcv closed 3 months ago

aqingcv commented 4 months ago

For V-COCO, you use the utilities provided by Gupta et al, but the length of the test dataset is 4532 in your code，this means that the resulting cache.pkl contains only 4532 outputs. However, in the utilities provided by Gupta et al., the images in the v-coco/data/splits/vcoco_test.ids is 4946, which is not equal to the 4532. Will this affect the final test results? Thanks.

fredzzhang commented 4 months ago

Hi @aqingcv,

I vaguely remember that I removed those images without valid human-object interactions, since some annotations in V-COCO only have <subject, predicate> pairs, but not the triplets.

Fred.