English(EN) Texture Representations in Deep Vision Models: Comparing CNNs, Vision Transformers, and Human Perception

研究发现视觉Transformer比卷积神经网络更能模拟人类纹理感知

作者 PulseAugur 编辑部 · [2 个来源] · 2026-07-09 10:05

一篇新的arXiv论文由Ludovica De Paolis撰写，比较了卷积神经网络（CNNs）和视觉Transformer（ViTs）如何表示纹理，这是视觉感知的一个关键方面。研究发现，ViTs创建的纹理表示与其复杂性无关，并且在识别纹理方面，人类的表现比CNNs更能被ViT表示所预测。这些发现表明，ViTs可能更准确地模拟人类纹理处理，而网络架构起着重要作用。 AI

影响表明视觉Transformer可能比传统的卷积神经网络提供更准确的人类纹理感知计算模型。

排序理由该集群包含一篇在arXiv上发表的研究论文，比较了不同的AI模型架构。

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.CV TIER_1 English(EN) · Ludovica de Paolis, Marco Baroni, Alessandro Laio, Eugenio Piasini · 2026-07-10 04:00

Texture Representations in Deep Vision Models: Comparing CNNs, Vision Transformers, and Human Perception

arXiv:2607.08321v1 Announce Type: new Abstract: In computational vision science, Convolutional Neural Networks (CNNs) have emerged as a popular model of biological vision because of the alignment they can exhibit with neural and behavioral data in humans and animals. However, it …
arXiv cs.CV TIER_1 English(EN) · Eugenio Piasini · 2026-07-09 10:05

Texture Representations in Deep Vision Models: Comparing CNNs, Vision Transformers, and Human Perception

In computational vision science, Convolutional Neural Networks (CNNs) have emerged as a popular model of biological vision because of the alignment they can exhibit with neural and behavioral data in humans and animals. However, it remains unclear to what extent this alignment pe…

报道来源 [2]

Texture Representations in Deep Vision Models: Comparing CNNs, Vision Transformers, and Human Perception

Texture Representations in Deep Vision Models: Comparing CNNs, Vision Transformers, and Human Perception

相关实体

相关话题