Low-Shot Learning from Imaginary 3D Model

TL;DR

This paper introduces a novel low-shot learning method that uses a 3D model to generate diverse training images from scarce samples, improving classification accuracy in few-shot scenarios.

Contribution

It proposes a 3D model-based image hallucination technique combined with self-paced learning for effective few-shot classification.

Findings

Significant accuracy improvement on CUB-200-2011 dataset

Effective image augmentation via 3D model hallucination

Enhanced classifier training with diverse high-quality images

Abstract

Since the advent of deep learning, neural networks have demonstrated remarkable results in many visual recognition tasks, constantly pushing the limits. However, the state-of-the-art approaches are largely unsuitable in scarce data regimes. To address this shortcoming, this paper proposes employing a 3D model, which is derived from training images. Such a model can then be used to hallucinate novel viewpoints and poses for the scarce samples of the few-shot learning scenario. A self-paced learning approach allows for the selection of a diverse set of high-quality images, which facilitates the training of a classifier. The performance of the proposed approach is showcased on the fine-grained CUB-200-2011 dataset in a few-shot setting and significantly improves our baseline accuracy.