Oscar, VinVL Pre-training: Bugs and Solutions
2021-08-18 09:33:24
Paper:
Oscar: Object-Semantics Aligned Pre-training for Vision-and-Language Tasks, ECCV 2020,
VinVL: Revisiting Visual Representations in Vision-Language Models, arxiv 2021
Code: VinVL-Doc, SceneGraphBenchmark, Oscar
1.
Oscar, VinVL Pre-training: Bugs and Solutions
原文：https://www.cnblogs.com/wangxiaocvpr/p/15155409.html