提出稳定对比学习方法，让像素同时保留语义与几何信息
Stable Single-Pixel Contrastive Learning for Semantic and Geometric Tasks
- 用过完备描述符表示像素，实现视角不变且语义明确
- 无需动量教师-学生训练即可实现精准跨图像点对应
- 在2D/3D合成环境验证了表示质量与稳定性
我们提出一类用于学习像素级表征的稳定对比损失，可联合捕捉语义与几何信息。该方法将图像中每个像素映射为一个过完备描述符，兼具视角不变性和语义意义。无需动量教师-学生训练，即可实现跨图像的精确点对应。在合成2D和3D环境中进行的两项实验验证了该损失及其生成的过完备表征的特性。
原文摘要 · Abstract (English)
We pilot a family of stable contrastive losses for learning pixel-level representations that jointly capture semantic and geometric information. Our approach maps each pixel of an image to an overcomplete descriptor that is both view-invariant and semantically meaningful. It enables precise point-correspondence across images without requiring momentum-based teacher-student training. Two experiments in synthetic 2D and 3D environments demonstrate the properties of our loss and the resulting overcomplete representations.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。