发表机构
Information & Electronics Research Institute, KAIST; Zoox Inc.; KAIST (Korea Advanced Institute of Science and Technology); Hanwha Aerospace; School of Electrical Engineering, KAIST（韩国科学技术院信息与电子研究院; Zoox公司; 韩国科学技术院; 韩华宇航; 韩国科学技术院电气工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对语义分割中上下文捕捉与长尾分布问题，提出含CCL、BANE采样的Contextrast++，在无额外推理开销下提升了基于对比学习的SOTA方法性能。
AI 中文摘要
语义分割借助深度学习已取得快速进展，但在有效捕捉局部与全局上下文、解决长尾分布问题方面仍存在挑战。为应对这些问题，本文提出Contextrast++，一种用于语义分割的鲁棒对比学习方法，可改进多尺度特征集成并缓解类别不平衡问题。该方法包含两个关键组件：1）上下文对比学习（CCL）；2）边界感知负样本（BANE）采样。CCL包含三个子组件：自适应融合模块、像素到锚点（PA）损失、锚点到锚点（AA）损失。自适应融合模块动态平衡局部与全局特征集成，生成更具上下文感知的表示。PA损失利用融合后的多尺度特征改进特征表示学习，AA损失则通过存储固定数量类别平衡代表性锚点的记忆库，聚焦解决长尾分布问题。同时，BANE采样通过从误分类边界区域选择难负样本，提升分割精度，在对比学习过程中优化细粒度细节。经公开数据集上的大量实验验证，Contextrast++相比现有基于对比学习的语义分割SOTA方法显著提升性能，且推理阶段无额外计算开销。
英文摘要
Semantic segmentation has rapidly advanced with deep learning; however, challenges remain in effectively capturing local and global contexts as well as addressing the long-tailed distribution problem. To tackle these issues, we present Contextrast++, a robust contrastive learning method for semantic segmentation that improves multi-scale feature integration and mitigates class imbalance issues. Our method consists of two key components: 1) contextual contrastive learning (CCL) and 2) boundary-aware negative (BANE) sampling. CCL includes three subcomponents: adaptive fusion module, pixel-to-anchor (PA) loss, and anchor-to-anchor (AA) loss. The adaptive fusion module dynamically balances local and global feature integration, resulting in a more context-aware representation. While the PA loss leverages the fused multi-scale features to improve feature representation learning, the AA loss focuses on addressing the long-tailed distribution problem by utilizing a memory bank that stores a fixed number of class-balanced representative anchors. Meanwhile, BANE sampling enhances segmentation precision by selecting hard negatives from misclassified boundary regions, which refines fine-grained details during contrastive learning. As verified in extensive experiments using public datasets, we demonstrate that Contextrast++ substantially improves semantic segmentation performance over existing contrastive learning-based state-of-the-art approaches, while introducing no additional computational overhead during inference.
CommentsAccepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2026
