AI Paper Atlas

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

Large-Scale Object Detection in the Wild From Imbalanced Multi-Labels

Junran Peng, Xingyuan Bu, Ming Sun, Zhaoxiang Zhang, Tieniu Tan, Junjie Yan

CVPR 2020 · Detection and segmentation › Open-set and few-shot detection · approach: Open-set and unknown object detection · 64 citations (OpenAlex)

Check 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
Use concurrent softmax instead of softmax loss during training (ablation)Open Images detection with class-aware sampling · Open Images mini-valmAP: 55.45 -> 56.58Improved
Use concurrent softmax instead of softmax loss during training (ablation)Open Images detection with non-balance sampling · Open Images mini-valmAP: 38.16 -> 39.14Improved
Apply concurrent softmax at inference only (ablation)Softmax-trained detector · Open Images mini-valmAP: 55.45 -> 55.77Improved
Use BCE or focal loss rather than softmax-based losses (ablation)Open Images multi-label detection · Open Images mini-valmAP: 50.18; 54.29Hurt
Use soft-balance sampling with λ=0.7 rather than class-aware sampling (ablation)Open Images detection · Open Images mini-valmAP: 55.45 -> 57.04Improved
Set soft-balance λ to 0.3 (ablation)Open Images detection · Open Images mini-valmAP: 50.69Hurt
Set soft-balance λ to 1.5 (ablation)Open Images detection · Open Images mini-valmAP: 52.41Hurt
Extend class-aware training from 7 to 14 epochs with ImageNet pretraining (ablation)ResNeXt-152 detector · Open Images full-valmAP: 64.68 -> 62.85Hurt
Add class-aware finetuning after non-balance training (ablation)ResNeXt-152 detector initialized from ImageNet · Open Images full-valmAP: 59.85 -> 65.60Improved
Add class-aware finetuning after non-balance training from scratch (ablation)ResNeXt-152 detector · Open Images full-valmAP: 60.70 -> 65.92Improved
Use soft-balance rather than class-aware finetuning after non-balance training from scratch (ablation)ResNeXt-152 detector · Open Images full-valmAP: 65.92 -> 67.09Improved
Apply concurrent softmax in training and testing (ablation)Soft-balance hybrid ResNeXt-152 detector · Open Images full-valmAP: 67.09 -> 68.23Improved

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