Why Do Line Drawings Work? A Realism Hypothesis

The Role of Edges in Line Drawing Perception

It has often been conjectured that the effectiveness of line drawings can be explained by the similarity of edge images to line drawings. This paper presents several problems with explaining line drawing perception in terms of edges, and how the recently-proposed Realism Hypothesis of Hertzmann (2020) resolves these problems. There is nonetheless existing evidence that edges are often the best features for predicting where people draw lines; this paper describes how the Realism Hypothesis can explain this evidence.

LineArt：一种基于知识引导的免训练高质量外观迁移方法，用于设计图纸的扩散模型

LineArt: A Knowledge-guided Training-free High-quality Appearance Transfer for Design Drawing with Diffusion Model

LineArt提出一种免训练、知识引导的外观迁移框架，通过多频率线条融合和两阶段绘画，在保持结构准确的同时提升设计图纸渲染的精度与真实感，并引入ProLines数据集验证其优于现有方法。

Physion-Eval：通过人类推理评估生成视频的物理真实性

Physion-Eval: Evaluating Physical Realism in Generated Video via Human Reasoning

本文提出Physion-Eval基准，通过专家人类推理评估生成视频的物理真实性，揭示当前视频生成模型在物理关键场景中存在显著缺陷，83.3%的外向视角和93.5%的内向视角视频存在可识别的物理错误。

为什么视觉语言模型会漏检小目标，以及何时放大是安全的

Why VLMs Miss Small Objects, and When Zooming In Is Safe

本研究构建理论解释VLM漏检小目标的原因，证明token预算与覆盖成本的权衡，并验证局部分解在特定条件下不降低召回率，且能提升性能。