[CVPR19] RAVEN: A Dataset for Relational and Analogical Visual Reasoning
(a) An example Raven Progressive Matrix. One is asked to select an image that best completes the problem matrix, following the structural and analogical relations. Each image has an underlying structure. (b) Specifically in this problem, it is an inside-outside structure in which the outside component is a layout with a single centered object and the inside component is a $2 \times 2$ grid layout. (c ) lists the rules for (a). The compositional nature of the rules makes this problem a difficult one. The correct answer is 7.Abstract
Dramatic progress has been witnessed in basic vision tasks involving low-level perception, such as object recognition, detection, and tracking. Unfortunately, there is still an enormous performance gap between artificial vision systems and human intelligence in terms of higher-level vision problems, especially ones involving reasoning. Earlier attempts in equipping machines with high-level reasoning have hovered around Visual Question Answering (VQA), one typical task associating vision and language understanding. In this work, we propose a new dataset, built in the context of Raven’s Progressive Matrices (RPM) and aimed at lifting machine intelligence by associating vision with structural, relational, and analogical reasoning in a hierarchical representation. Unlike previous works in measuring abstract reasoning using RPM, we establish a semantic link between vision and reasoning by providing structure representation. This addition enables a new type of abstract reasoning by jointly operating on the structure representation. Machine reasoning ability using modern computer vision is evaluated in this newly proposed dataset. Additionally, we also provide human performance as a reference. Finally, we show consistent improvement across all models by incorporating a simple neural module that combines visual understanding and structure reasoning.
Chi Zhang
Research Scientist
Feng Gao
Applied Scientist
Baoxiong Jia
Research Scientist
