Research map / Detection and segmentation / Open-set and few-shot detection
Described Object Detection: Liberating Object Detection with Flexible Expressions
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 granularity decomposition (separate decoder for local tasks) (ablation) | OFA · D3 intra FULL | mAP: 3.4 -> 10.5 | Helped |
| add reconstructed data (multi-reference, zero-or-many targets) (ablation) | OFA+GD · D3 intra FULL | mAP: 10.5 -> 17.2 | Helped |
| add task decomposition (REC then binary VQA verification) (ablation) | OFA+GD+RD · D3 intra FULL | mAP: 17.2 -> 21.6 | Helped |
| remove detection data from training (ablation) | OFA-DOD · D3 intra FULL | mAP: 21.6 -> 16.4 | Hurt |
| remove image-to-text data from training (ablation) | OFA-DOD · D3 intra FULL | mAP: 21.6 -> 14.2 | Hurt |
| remove MLM data from training (ablation) | OFA-DOD · D3 intra FULL | mAP: 21.6 -> 20.3 | No clear effect |
| use inter-scenario evaluation instead of intra-scenario | REC method OFA · D3 | mAP: 4.2 -> 0.1 | Hurt |
| increase description length from short to very long | OWL-ViT (OVD) · D3 | mAP: 20.7 -> 5.3 | Hurt |
| increase description length from short to very long | OFA-DOD · D3 | mAP: 23.6 -> 18.4 | Hurt |
| absence instead of presence descriptions | OFA-DOD · D3 intra | mAP: 23.7 -> 15.4 | Hurt |
| bi-functional (REC+OVD) training vs OVD-only | Grounding-DINO vs OWL-ViT · D3 intra FULL | mAP: 9.6 -> 20.7 | Helped |
Closest papers by meaning
- Generative Region-Language Pretraining for Open-Ended Object Detection (CVPR 2024 · similarity 0.91)
- How to Evaluate the Generalization of Detection? A Benchmark for Comprehensive Open-Vocabulary Detection (AAAI 2024 · similarity 0.91)
- The Devil is in the Fine-Grained Details: Evaluating Open-Vocabulary Object Detectors for Fine-Grained Understanding (CVPR 2024 · similarity 0.91)
- Revisiting Few-Shot Object Detection with Vision-Language Models (NeurIPS 2024 · similarity 0.90)
- DoGA: Enhancing Grounded Object Detection via Grouped Pre-Training with Attributes (AAAI 2025 · similarity 0.90)
- Roboflow100-VL: A Multi-Domain Object Detection Benchmark for Vision-Language Models (NeurIPS 2025 · 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.