Research map / Detection and segmentation / Open-set and few-shot detection
Towards Open Vocabulary Learning: A Survey
Jianzong Wu, Xiangtai Li, Shilin Xu, Haobo Yuan, Henghui Ding, Yibo Yang, Xia Li, Jiangning Zhang, Yunhai Tong, Xudong Jiang, Bernard Ghanem, Dacheng Tao
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 |
|---|---|---|---|
| scale backbone from RN50-FPN to RN50x64 | F-VLM · LVIS 1.0 open-vocab detection | mask AP_r: 18.6 -> 32.8 | Improved |
| scale backbone from RN50 to RN50x64 | DST-Det · V3Det open-vocab detection | AP novel: 7.2 -> 13.5 | Improved |
| scale backbone from ViT-B/16 to ViT-L/16 | CLIPSelf · LVIS 1.0 open-vocab detection | mask AP_r: 25.3 -> 34.9 | Improved |
| replace RN50-FPN with Swin-B backbone | Detic · LVIS 1.0 open-vocab detection | mask AP_r: 19.5 -> 23.9 | Improved |
| replace RN50-FPN with Swin-B backbone | VLDet · LVIS 1.0 open-vocab detection | mask AP_r: 21.7 -> 26.3 | Improved |
| scale ViT-B/16 to ViT-L/14 | Open-VCLIP · UCF101 open-vocab video classification | accuracy: 83.4 -> 87.6 | Improved |
| replace R50 with SwinB backbone | MindVLT · LV-VIS val | mAP_all: 14.1 -> 21.4 | Improved |
| use MindVLT instead of Detic-OWTB tracking | Detic-OWTB (R50) · LV-VIS val | mAP novel: 5.3 -> 9.4 | Improved |
| add Objaverse-LVIS to OpenShape training data | OpenShape-PointBERT · Objaverse-LVIS zero-shot 3D classification | Top1 accuracy: 39.1 -> 46.8 | Improved |
| train on OpenShape ensembled data instead of ShapeNet only | OpenShape-PointBERT · Objaverse-LVIS zero-shot 3D classification | Top1 accuracy: 10.8 -> 39.1 | Improved |
| add language supervision (PLA training) | PLA w/o training · ScanNet open-world 3D seg (B15/N4) | hIoU: 39.7 -> 65.3 | Improved |
| replace PLA with RegionPLC (regional language supervision) | PLA · ScanNet open-world 3D seg (B15/N4) | hIoU: 65.3 -> 69.9 | Improved |
Closest papers by meaning
- A Survey on Open-Vocabulary Detection and Segmentation: Past, Present, and Future (TPAMI 2024 · similarity 0.95)
- A Simple Baseline for Open-Vocabulary Semantic Segmentation with Pre-trained Vision-Language Model (ECCV 2022 · similarity 0.93)
- Emergent Open-Vocabulary Semantic Segmentation from Off-the-shelf Vision-Language Models (CVPR 2024 · similarity 0.93)
- Boosting Segment Anything Model Towards Open-Vocabulary Learning (AAAI 2025 · similarity 0.91)
- Transferable and Principled Efficiency for Open-Vocabulary Segmentation (CVPR 2024 · similarity 0.90)
- LLMs Meet VLMs: Boost Open Vocabulary Object Detection with Fine-grained Descriptors (ICLR 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.
