arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.07991cs.CVcs.LGeess.IVq-bio.NC
收藏

视觉注视感知系统的涌现性质

Emergent Properties of Foveated Perceptual Systems

Center for Brains, Minds and Machines(脑、心智与机器中心);Massachusetts Institute of Technology(麻省理工学院);Harvard University(哈佛大学)

Arturo Deza, Talia Konkle

摘要:

这项工作的目标是表征的代表性的影响,中央凹操作的机器视觉系统,受到中央凹的人类视觉系统,它具有较高的敏锐度在中心的凝视和纹理般的编码在外围的启发。为此,我们引入了由第一阶段\textit{fixed}图像变换和第二阶段\textit{learnable}卷积神经网络组成的模型,并且我们改变了第一阶段组件。主模型具有一个中心凹纹理输入阶段,我们将其与具有中心凹模糊输入的模型和具有空间均匀模糊输入的模型(两者都匹配感知压缩)以及具有最小输入压缩的最终参考模型进行比较。我们发现:1)尽管其压缩输入,但中心凹纹理模型显示出与参考模型相似的场景分类精度,具有更大的i.i.d.与其他模型相比,中心凹纹理模型具有更高的泛化能力; 2)与比较模型相比,中心凹纹理模型对高空间频率信息具有更高的敏感性,并且对遮挡具有更大的鲁棒性; 3)即使具有权重共享约束,两个中心凹系统相对于空间均匀系统也显示出更强的中心图像偏置。至关重要的是,这些结果在不同的经典CNN架构的整个学习动态中得到了保留。总而言之,这表明,视觉聚焦与基于外围纹理的计算产生了一个有效的,独特的,和强大的场景信息的表示格式,并提供共生的计算洞察到表示的后果,基于纹理的外围编码可能具有在人类视觉系统中的处理,同时还可能激发下一代的计算机视觉模型通过空间自适应计算。代码+数据可在这里:https://github.com/ArturoDeza/EmergentProperties

英文摘要:

The goal of this work is to characterize the representational impact that foveation operations have for machine vision systems, inspired by the foveated human visual system, which has higher acuity at the center of gaze and texture-like encoding in the periphery. To do so, we introduce models consisting of a first-stage \textit{fixed} image transform followed by a second-stage \textit{learnable} convolutional neural network, and we varied the first stage component. The primary model has a foveated-textural input stage, which we compare to a model with foveated-blurred input and a model with spatially-uniform blurred input (both matched for perceptual compression), and a final reference model with minimal input-based compression. We find that: 1) the foveated-texture model shows similar scene classification accuracy as the reference model despite its compressed input, with greater i.i.d. generalization than the other models; 2) the foveated-texture model has greater sensitivity to high-spatial frequency information and greater robustness to occlusion, w.r.t the comparison models; 3) both the foveated systems, show a stronger center image-bias relative to the spatially-uniform systems even with a weight sharing constraint. Critically, these results are preserved over different classical CNN architectures throughout their learning dynamics. Altogether, this suggests that foveation with peripheral texture-based computations yields an efficient, distinct, and robust representational format of scene information, and provides symbiotic computational insight into the representational consequences that texture-based peripheral encoding may have for processing in the human visual system, while also potentially inspiring the next generation of computer vision models via spatially-adaptive computation. Code + Data available here: https://github.com/ArturoDeza/EmergentProperties

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。

↑