AI Paper Atlas

Research map / Detection and segmentation / CNN features

MutDet: Mutually Optimizing Pre-training for Remote Sensing Object Detection

Ziyue Huang, Yongchao Feng, Qingjie Liu, Yunhong Wang

ECCV 2024 · Detection and segmentation › CNN features · approach: Box regression and loss design · 11 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
add mutual enhancement module feeding enhanced features to decoder (ablation)MutDet · DIOR-R 12 epochsAP50: 62.8 -> 65.5Improved
add auxiliary siamese head for calibration instead of distillation (ablation)MutDet · DIOR-R (pretrain DOTA-v1.0)AP50: 69.4 -> 70.7Improved
calibrate with encoder feature distillation (ablation)MutDet · DIOR-R (pretrain DOTA-v1.0)AP50: 69.4 -> 69.6No clear effect
calibrate with decoder cross distillation (ablation)MutDet · DIOR-R (pretrain DOTA-v1.0)AP50: 69.4 -> 69.9Improved
add contrastive alignment loss to DETReg baseline (ablation)DETReg · DIOR-R 36 epochsAP50: 67.9 -> 68.5Improved
add encoder contrastive alignment loss without siamese head (ablation)MutDet · DIOR-R 36 epochsAP50: 69.5 -> 69.4No clear effect
use MutDet pre-training instead of DETRegARS-DETR R50 · DIOR-R 10% labeled dataAP50: 50.8 -> 56.9Improved
pre-train on DOTA then fine-tune on DOTA with prior methodsUP-DETR, AlignDet, DETReg · DOTA-v1.0 testAP50: 72.9 -> 72.6Hurt
use stronger Swin-Tiny backbone and larger RSDet4 pre-training dataMutDet vs DETReg · DIOR-R 36 epochsAP50: 73.2 -> 73.7No clear effect
use MutDet pre-training instead of DETReg on other DETR detectorsRotated D-DETR · DIOR-RAP50: 41.4 -> 45.8Improved
AlignDet pre-training instead of noneARS-DETR R50 · DIOR-R 12 epochsAP50: 55.4 -> 54.3Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

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