English

Spatial Knowledge Distillation to aid Visual Reasoning

Computer Vision and Pattern Recognition 2018-12-12 v2

Abstract

For tasks involving language and vision, the current state-of-the-art methods tend not to leverage any additional information that might be present to gather relevant (commonsense) knowledge. A representative task is Visual Question Answering where large diagnostic datasets have been proposed to test a system's capability of answering questions about images. The training data is often accompanied by annotations of individual object properties and spatial locations. In this work, we take a step towards integrating this additional privileged information in the form of spatial knowledge to aid in visual reasoning. We propose a framework that combines recent advances in knowledge distillation (teacher-student framework), relational reasoning and probabilistic logical languages to incorporate such knowledge in existing neural networks for the task of Visual Question Answering. Specifically, for a question posed against an image, we use a probabilistic logical language to encode the spatial knowledge and the spatial understanding about the question in the form of a mask that is directly provided to the teacher network. The student network learns from the ground-truth information as well as the teachers prediction via distillation. We also demonstrate the impact of predicting such a mask inside the teachers network using attention. Empirically, we show that both the methods improve the test accuracy over a state-of-the-art approach on a publicly available dataset.

Keywords

Cite

@article{arxiv.1812.03631,
  title  = {Spatial Knowledge Distillation to aid Visual Reasoning},
  author = {Somak Aditya and Rudra Saha and Yezhou Yang and Chitta Baral},
  journal= {arXiv preprint arXiv:1812.03631},
  year   = {2018}
}

Comments

Equal contribution by first two authors. Accepted in WACV 2019

R2 v1 2026-06-23T06:37:04.876Z