Challenges and Prospects in Vision and Language Research
arXiv:1904.09317
Abstract
Language grounded image understanding tasks have often been proposed as a method for evaluating progress in artificial intelligence. Ideally, these tasks should test a plethora of capabilities that integrate computer vision, reasoning, and natural language understanding. However, rather than behaving as visual Turing tests, recent studies have demonstrated state-of-the-art systems are achieving good performance through flaws in datasets and evaluation procedures. We review the current state of affairs and outline a path forward.
References in corpus (10)
- Generative Adversarial Text to Image Synthesis
- The Natural Language Decathlon: Multitask Learning as Question Answering
- Visual Question Answering: Datasets, Algorithms, and Future Challenges
- Why do deep convolutional networks generalize so poorly to small image transformations?
- Exploring Nearest Neighbor Approaches for Image Captioning
- Blindfold Baselines for Embodied QA
- Visual Dialogue without Vision or Dialogue
- VQD: Visual Query Detection in Natural Scenes
- Do Explanations make VQA Models more Predictable to a Human?
- Visual Referring Expression Recognition: What Do Systems Actually Learn?