Research map / Detection and segmentation / Open-set and few-shot detection
Zero-Shot Object Detection with Textual Descriptions
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 pre-training of the network (ablation) | zero-shot detector with textual descriptions · ILSVRC-2017 | mAP: 12.2 -> 24.1 | Helped |
| add pre-training (ablation) | zero-shot detector with textual descriptions · MSCOCO IoU 0.5 | Recall@100: 21.2 -> 34.3 | Helped |
| add pre-training (ablation) | zero-shot detector with textual descriptions · Visual Genome IoU 0.5 | Recall@100: 1.1 -> 7.2 | Helped |
| add word-level attention (ablation) | zero-shot detector with textual descriptions · ILSVRC-2017 | mAP: 18.4 -> 24.1 | Helped |
| add word-level attention (ablation) | zero-shot detector with textual descriptions · MSCOCO IoU 0.5 | Recall@100: 28.1 -> 34.3 | Helped |
| add word-level attention (ablation) | zero-shot detector with textual descriptions · Visual Genome IoU 0.5 | Recall@100: 3.2 -> 7.2 | Helped |
| add visual-unit attention instead of average pooling (ablation) | zero-shot detector with textual descriptions · ILSVRC-2017 | mAP: 9.2 -> 24.1 | Helped |
| add visual-unit attention instead of average pooling (ablation) | zero-shot detector with textual descriptions · MSCOCO IoU 0.5 | Recall@100: 15.1 -> 34.3 | Helped |
| use textual description embedding instead of concept-name word embedding in baseline | zero-shot detector with textual descriptions · LAB baseline, Visual Genome IoU 0.5 | Recall@100: 5.4 -> 7.2 | Helped |
| proposed model vs second-best baseline (DSES) | zero-shot detector with textual descriptions · ILSVRC-2017 | mAP: 22.7 -> 24.1 | Helped |
| proposed model vs DSES | zero-shot detector · Visual Genome IoU 0.4 | Recall@100: 8.6 -> 9.7 | Helped |
Closest papers by meaning
- Context-Aware Zero-Shot Recognition (AAAI 2020 · similarity 0.88)
- Revisiting Few-Shot Object Detection with Vision-Language Models (NeurIPS 2024 · similarity 0.87)
- Roboflow100-VL: A Multi-Domain Object Detection Benchmark for Vision-Language Models (NeurIPS 2025 · similarity 0.86)
- Improved Zero-Shot Classification by Adapting VLMs with Text Descriptions (CVPR 2024 · similarity 0.86)
- Zero-shot Generalizable Incremental Learning for Vision-Language Object Detection (NeurIPS 2024 · similarity 0.86)
- Boosting Segment Anything Model Towards Open-Vocabulary Learning (AAAI 2025 · similarity 0.86)
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.