PulseAugur
中
实时 07:24:47
English(EN) DOD-SA: Infrared-Visible Decoupled Object Detection with Single-Modality Annotations

新的DOD-SA框架降低了红外-可见光目标检测的标注成本

作者 PulseAugur 编辑部 · [1 个来源] ·

研究人员开发了一个名为DOD-SA的新框架，用于红外-可见光目标检测，旨在降低现有方法的高昂标注成本。该框架利用单模态和双模态协同师生网络（CoSD-TSNet）实现不同模态之间的知识迁移。该系统采用渐进式和自适应训练策略（PaST）以及伪标签分配器（PLA）来提高伪标签的准确性并处理训练过程中的模态失准问题。 AI

排序理由 该集群包含一篇详细介绍新目标检测技术框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

新的DOD-SA框架降低了红外-可见光目标检测的标注成本

本文如何被排名

Signal score
22 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该集群包含一篇详细介绍新目标检测技术框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, other
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

  1. arXiv cs.CV TIER_1 English(EN) · Hang Jin, Chenqiang Gao, Junjie Guo, Fangcen Liu, Qinyao Chang, Kanghui Tian, Deyu Meng ·

    DOD-SA：单模态标注下的红外-可见光解耦目标检测

    arXiv:2508.10445v2 Announce Type: replace Abstract: Infrared-visible object detection has shown great potential in real-world applications, enabling robust all-day perception by leveraging the complementary information of infrared and visible images. However, existing methods typ…

相关实体

相关话题