CRIPP-VQA

Counterfactual Reasoning about Implicit Physical Properties via Video Question Answering

VideosMIT licenseIntroduced 2022-11-07

CRIPP-VQA is a video question answering dataset for reasoning about the implicit physical properties of objects in a scene. It contains videos of object in motion, annotated with questions that involve counterfactual reasoning about actions, questions about planning in order to reach a goal, and descriptive questions about visible properties of objects.

Source: CRIPP-VQA: Counterfactual Reasoning about Implicit Physical Properties via Video Question Answering

Image Source: https://arxiv.org/pdf/2211.03779v1.pdf