There are 1 repository under conceptual-captions topic.
Cross-lingual Visual Pre-training for Multimodal Machine Translation
Preserving Semantic Neighborhoods for Robust Cross-modal Retrieval [ECCV 2020]
A modular repository for developing Image Captioning Approaches
The main goal of is to show how precise the Faster R-CNN with ResNet-101 could find objects and there attributes in Conceptual 12m dataset.