The Ikshana Hypothesis of Human Scene Understanding
This work addresses the data efficiency problem in computer vision for researchers and practitioners, but it is incremental as it builds on existing neuroscience insights.
The authors tackled the problem of deep neural networks requiring massive labeled data for computer vision by proposing the Ikshana hypothesis, inspired by neuroscience, and designed IkshanaNet, which outperformed baselines on Cityscapes and CamVid semantic segmentation benchmarks.
In recent years, deep neural networks (DNNs) achieved state-of-the-art performance on several computer vision tasks. However, the one typical drawback of these DNNs is the requirement of massive labeled data. Even though few-shot learning methods address this problem, they often use techniques such as meta-learning and metric-learning on top of the existing methods. In this work, we address this problem from a neuroscience perspective by proposing a hypothesis named Ikshana, which is supported by several findings in neuroscience. Our hypothesis approximates the refining process of conceptual gist in the human brain while understanding a natural scene/image. While our hypothesis holds no particular novelty in neuroscience, it provides a novel perspective for designing DNNs for vision tasks. By following the Ikshana hypothesis, we design a novel neural-inspired CNN architecture named IkshanaNet. The empirical results demonstrate the effectiveness of our method by outperforming several baselines on the entire and subsets of the Cityscapes and the CamVid semantic segmentation benchmarks.