Research map / Detection and segmentation / Open-set and few-shot detection
NoOVD: Novel Category Discovery and Embedding for Open-Vocabulary Object Detection
Yupeng Zhang, Ruize Han, Zhiwei Chen, Wei Feng, Liang Wan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add NoOVD (latent novel discovery + self-distillation + R-RPN) | CLIPSelf + F-ViT ViT-B/16 · OV-LVIS | APr: 25.4 -> 28.3 | Improved |
| add NoOVD | DeCLIP + F-ViT ViT-L/14 · OV-LVIS | APr: 36.9 -> 39.2 | Improved |
| add NoOVD | CLIPSelf + F-ViT ViT-B/16 · OV-COCO | AP50 novel: 35.8 -> 37.9 | Improved |
| add NoOVD | CLIPSelf + F-ViT ViT-L/14 · OV-COCO | AP50 base: 64.8 -> 64.7 | No clear effect |
| add NoOVD | DeCLIP + F-ViT ViT-L/14 · Objects365 (cross-dataset) | APr: 22.3 -> 23.3 | Improved |
| add K-FPN features instead of CLIP-top (ablation) | NoOVD ViT-B/16 · OV-LVIS | APr: 26.4 -> 27.5 | Improved |
| add R-RPN at test time (ablation) | F-ViT baseline ViT-B/16 · OV-LVIS | APr: 25.4 -> 26.7 | Improved |
| use CLIP top-layer feature for discovery (ablation) | F-ViT baseline ViT-B/16 · OV-LVIS | APr: 25.4 -> 26.4 | Improved |
| tune K-FPN fusion weight W from 0.8 to 0.3 (ablation) | K-FPN · OV-LVIS (CLIPSelf ViT-B/16) | AP: 19.1 -> 26.7 | Improved |
| set R-RPN alpha 0.5 vs 0.9 (ablation) | R-RPN · OV-LVIS (CLIPSelf ViT-B/16) | AP: 25.5 -> 26.7 | Improved |
| use L2 instead of cosine loss for distillation (ablation) | NoOVD · OV-LVIS (CLIPSelf ViT-B/16) | APr: 17.3 -> 28.3 | Improved |
| use L2 instead of L1/SmoothL1 (ablation) | NoOVD · OV-LVIS (CLIPSelf ViT-B/16) | APr: 27.8/28.0 -> 28.3 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- OV-DQUO — OV-DQUO: Open-Vocabulary DETR with Denoising Text Query Training and Open-World Unknown Objects Supervision (AAAI 2025) · better on 0 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- CAKE: Category Aware Knowledge Extraction for Open-Vocabulary Object Detection (AAAI 2025 · similarity 0.93)
- Cyclic Contrastive Knowledge Transfer for Open-Vocabulary Object Detection (ICLR 2025 · similarity 0.91)
- ProxyDet: Synthesizing Proxy Novel Classes via Classwise Mixup for Open-Vocabulary Object Detection (AAAI 2024 · similarity 0.91)
- DitHub: A Modular Framework for Incremental Open-Vocabulary Object Detection (NeurIPS 2025 · similarity 0.91)
- Learning Background Prompts to Discover Implicit Knowledge for Open Vocabulary Object Detection (CVPR 2024 · similarity 0.91)
- SHiNe: Semantic Hierarchy Nexus for Open-vocabulary Object Detection (CVPR 2024 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
