---
格式版本: 2
标题: "OV3D-Bench: A Diagnostic Benchmark for Open-Vocabulary Monocular 3D Detection"
原文链接: "https://arxiv.org/abs/2608.17110"
发布日期: "2026-08-17"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "rule:local:strict_original_body"
发布时间证据: "**\\[v1\\]** Mon, 17 Aug 2026 20:38:57 UTC (3,283 KB)"
发布时间校准原因: "规则确认唯一严格发布时间，来源 local:strict_original_body"
发布时间校准置信度: "high"
发布时间候选数量: 18
发布时间严格候选数量: 6
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-08-20T14:44:18+08:00"
发布时间仲裁状态: "skipped"
发布时间仲裁尝试次数: 0
发布时间仲裁耗时毫秒: 0
发现时间: "2026-08-20T14:38:42+08:00"
入库时间: "2026-08-20T06:44:18.438Z"
来源平台: "arXiv 学术论文搜索"
搜索渠道: "source_template"
搜索词: "https://arxiv.org/search/?query=Oracle&searchtype=all"
匹配关键词:
  - "deployment"
  - "performance"
相关厂家:
  - "Oracle"
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI优质: "否"
AI打分: 2
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "正文为计算机视觉领域开放词汇3D检测论文，与超节点、AI Rack、机柜级AI基础设施等主题完全无关，且未命中任何相关关键词，属于明显无关内容。"
AI质检模型: "ali-deepseek-v4-flash"
AI质检时间: "2026-08-20T14:47:27+08:00"
AI主题相关性: 0
AI来源权威性: 1
AI新颖性: 0
AI技术细节: 0
AI商业部署信号: 0
AI完整性: 1
AI摘要: "OV3D-Bench 在七个室内外数据集上评估七种开放词汇单目3D检测器，发现模型定位准确但常将正确框误标为语义相邻类别，且结果对提示词措辞极敏感，如 WildDet3D 在描述性提示下 AP 从18.6降至5.4；"
AI摘要模型: "ali-deepseek-v4-flash"
AI摘要时间: "2026-09-07T03:25:28.927Z"
采集批次: "2026年8月20日14点19分32秒"
采集批次ID: "20260820-141932-079"
去重键: "https://arxiv.org/abs/2608.17110"
---

## Computer Science > Computer Vision and Pattern Recognition

## Title:OV3D-Bench: A Diagnostic Benchmark for Open-Vocabulary Monocular 3D Detection

Authors:[Mariia Gladkova](https://arxiv.org/search/cs?searchtype=author&query=Gladkova,+M), [Neehar Peri](https://arxiv.org/search/cs?searchtype=author&query=Peri,+N), [Ishan Khatri](https://arxiv.org/search/cs?searchtype=author&query=Khatri,+I), [Deva Ramanan](https://arxiv.org/search/cs?searchtype=author&query=Ramanan,+D), [Daniel Cremers](https://arxiv.org/search/cs?searchtype=author&query=Cremers,+D)

[View PDF](https://arxiv.org/pdf/2608.17110) [HTML (experimental)](https://arxiv.org/html/2608.17110v1)

> Abstract:Open-vocabulary monocular 3D detectors report strong in-domain performance, but each evaluates under a different protocol, several rely on per-image category oracles unavailable at deployment, and all collapse geometry and semantics into a single AP metric. To address this, we introduce OV3D-Bench, a diagnostic benchmark that compares open-vocabulary monocular 3D detectors under deployment-realistic conditions across seven indoor and outdoor datasets. Our benchmark replaces the per-image class name oracle with test-time dataset-level class name prompts, and decouples detection accuracy along three axes: localization, semantic robustness, and cross-domain transfer. We evaluate seven representative detectors and find that (i) they localize objects well yet often mislabel a correctly localized box as a semantically adjacent category; (ii) accuracy is highly sensitive to prompt phrasing (e.g. WildDet3D's performance collapses from 18.6 to 5.4 AP when prompted with "a detailed high-resolution photo of a car" rather than "car"); and (iii) the widely adopted target-aware protocol hides these errors (e.g. inflating DetAny3D's AP by 1.9 $\times$ on ScanNet). Lastly, we demonstrate that simply remapping a frozen closed-vocabulary detector's predictions using a contrastive vision-language encoder such as SigLIPv2 performs competitively against recent purpose-built open-vocabulary methods. This indicates that geometric localization is more mature, while open-vocabulary semantics remains the primary bottleneck.

| Comments: |  |
| --- | --- |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | [arXiv:2608.17110](https://arxiv.org/abs/2608.17110) \[cs.CV\] |
|  | (or [arXiv:2608.17110v1](https://arxiv.org/abs/2608.17110v1) \[cs.CV\] for this version) |
|  | [https://doi.org/10.48550/arXiv.2608.17110](https://doi.org/10.48550/arXiv.2608.17110) |

## Submission history

From: Mariia Gladkova \[[view email](https://arxiv.org/show-email/c98d673b/2608.17110)\]  
**\[v1\]** Mon, 17 Aug 2026 20:38:57 UTC (3,283 KB)

[Which authors of this paper are endorsers?](https://arxiv.org/auth/show-endorsers/2608.17110) | Disable MathJax ([What is MathJax?](https://info.arxiv.org/help/mathjax.html))