面向少样本概念学习的贝叶斯逆图形学
Bayesian Inverse Graphics for Few-Shot Concept Learning
AI总结:
本研究提出一种贝叶斯逆图形学模型,仅需少量样本即可学习对象概念,通过可微渲染、MCMC采样及神经似然函数推断物理参数,在少样本分类等任务上性能优于纯神经方法,具备强泛化性与不确定性感知能力。
摘要:
人类擅长仅从单个示例中构建对新概念的泛化认知。与之相反,当前的计算机视觉模型通常需要大量训练样本才能达到相当的准确率。在本研究中,我们提出了一种仅需极少数据即可学习的贝叶斯感知模型,即一种对象的原型化概率程序。具体而言,我们提出了一种针对基本形状的生成式逆图形学模型,可从一张或多张图像中推断出物理一致参数的后验分布。我们展示了该表示可用于少样本分类、姿态估计等下游任务。我们的模型性能优于现有的纯神经网络少样本分类算法,并展现出在不同光照条件、背景以及分布外形状下的泛化能力。通过设计,该模型具备不确定性感知能力,采用我们新提出的可微渲染器通过梯度下降优化全局场景参数,借助马尔可夫链蒙特卡洛(MCMC)对对象参数的后验分布进行采样,并使用基于神经网络的似然函数。
英文摘要:
Humans excel at building generalizations of new concepts from just one single example. Contrary to this, current computer vision models typically require large amount of training samples to achieve a comparable accuracy. In this work we present a Bayesian model of perception that learns using only minimal data, a prototypical probabilistic program of an object. Specifically, we propose a generative inverse graphics model of primitive shapes, to infer posterior distributions over physically consistent parameters from one or several images. We show how this representation can be used for downstream tasks such as few-shot classification and pose estimation. Our model outperforms existing few-shot neural-only classification algorithms and demonstrates generalization across varying lighting conditions, backgrounds, and out-of-distribution shapes. By design, our model is uncertainty-aware and uses our new differentiable renderer for optimizing global scene parameters through gradient descent, sampling posterior distributions over object parameters with Markov Chain Monte Carlo (MCMC), and using a neural based likelihood function.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。