AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

ScaleDet: A Scalable Multi-Dataset Object Detector

Yanbei Chen, Manchen Wang, Abhay Mittal, Zhenlin Xu, Paolo Favaro, Joseph Tighe, Davide Modolo

CVPR 2023 · Detection and segmentation › Open-set and few-shot detection · approach: Open-set and unknown object detection · 25 citations (OpenAlex)

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 changedWhereMeasuredEffect
Increase training data from one dataset to four datasetsScaleDet · LVISmAP: 33.1 -> 36.8Improved
Train on LVIS, COCO, Objects365, and OpenImages instead of training each dataset separatelyScaleDet · LVIS, COCO, Objects365, OpenImagesmAP: 33.1, 46.8, 28.8, 69.3 -> 36.8, 47.1, 30.6, 69.4Improved
Replace ResNet50 with Swin-BaseScaleDet · Upstream datasets (mean)mAP: 46.0 -> 56.3Improved
Use OpenCLIP instead of CLIP text embeddingsScaleDet-T · Upstream datasets (mean)mAP: 50.9 -> 51.4Improved
Use OpenCLIP instead of CLIP text embeddingsScaleDet-B · Upstream datasets (mean)mAP: 56.3 -> 56.5Improved
Fine-tune the multi-dataset model on each evaluation datasetScaleDet-B · Upstream datasets (mean)mAP: 56.5 -> 58.0Improved
Use one unified semantic label classifier rather than dataset-specific classifiersMulti-dataset detector · COCO, Objects365, OpenImagesmAP: 45.4 -> 47.7Improved
Train on four detection datasets rather than LVIS and COCO onlyScaleDet · LVIS, COCOmAP: 38.9 -> 41.9Improved
Use ScaleDet-B instead of GLIPv2-B for fine-tune transferObject detector · ODinW-13mAP: 69.4 -> 71.8Improved
Use ScaleDet-R instead of Detic-RObject detector · ODinW-13 direct transfermAP: 29.4 -> 39.4Improved
Add soft-label assignment loss to hard-label assignment (ablation)ScaleDet trained on L+C · LVIS, COCO, Objects365, OpenImagesmAP: 34.2 -> 34.5Improved
Add soft-label assignment loss to hard-label assignment (ablation)ScaleDet trained on L+C+O365+OID · ODinW-13 direct transfermAP: 37.9 -> 39.4Improved

Later papers that compared with this method

1 later accepted paper reports ScaleDet as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy