Why is [person4] pointing at [person1]?

Rationale: I think so because...

Visual Commonsense Reasoning (VCR) is a new task and large-scale dataset for cognition-level visual understanding.

With one glance at an image, we can effortlessly imagine the world beyond the pixels (e.g. that [person1] ordered pancakes). While this task is easy for humans, it is tremendously difficult for today's vision systems, requiring higher-order cognition and commonsense reasoning about the world. We formalize this task as Visual Commonsense Reasoning. In addition to answering challenging visual questions expressed in natural language, a model must provide a rationale explaining why its answer is true.

Overview of VCR

  • 290k multiple choice questions
  • 290k correct answers and rationales: one per question
  • 110k images
  • Counterfactual choices obtained with minimal bias, via our new Adversarial Matching approach
  • Answers are 7.5 words on average; rationales are 16 words.
  • High human agreement (>90%)
  • Scaffolded on top of 80 object categories from COCO
  • Is now (as of Dec 3) available for download!

From Recognition to Cognition: Visual Commonsense Reasoning

Authors

This paper was made possible by a group of researchers at the University of Washington and AI2. For announcements follow Rowan on twitter »

Sponsors