The Supplementary Material: Counterfactual Contrastive Learning for Weakly-Supervised Vision-Language Grounding
By Zhu Zhang, Zhou Zhao, Zhijie Lin, Jieming Zhu, Xiuqiang He. 2020
Subfields: Computer Vision and Pattern Recognition, Multimodal Learning
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Matt Gardner
- Sameer Singh
- Anna Rohrbach
- Trevor Darrell
- Luc Van Gool
- Xiaogang Wang
- Dacheng Tao
- Sanjay Subramanian
In Multimodal Learning
- Trevor Darrell
- Anna Rohrbach
- William Merrill
- Sanjay Subramanian
- Sameer Singh
- Matt Gardner
- Marcus Rohrbach
- Mohit Bansal
Citations in the corpus
- ReCLIP: A Strong Zero-Shot Baseline for Referring Expression Comprehension (2022, background): “There is a long line of work in weakly supervised ReC, where at training time, pairs of referring expressions and images are available but the ground-truth bounding boxes for each expression are not (Rohrbach et al., 2016; Liu et al., 2019; Zhang et al., 2018, 2020; Sun et al., 2021).”
- Making the Most of Text Semantics to Improve Biomedical Vision-Language Processing (2022, background): “Much of the general domain prior work on phrase grounding relies on off-the-shelf object-detection networks [7,12,27,75,83,86] such as Faster R-CNN [64] which are pretrained on large labelled datasets to extract region candidates from images.”