AI Paper Atlas

Research map / Detection and segmentation / CNN features

WB-DETR: Transformer-Based Detector Without Backbone

Fanfan Liu, Haoran Wei, Wenzhe Zhao, Guozhen Li, Jingquan Peng, Zihao Li

ICCV 2021 · Detection and segmentation › CNN features · approach: Region-based and query-based detectors · 43 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 LIE-T2T encoder layers (0 to 3 layers) (ablation)WB-DETR (2-8) · COCO val2017 (WB-DETR 2-8)AP: 30.7 -> 38.5 -> 40.2 -> 40.3Improved
add LIE-T2T layers, 0 to 2, for small objects (ablation)WB-DETR (2-8) · COCO val2017 (WB-DETR 2-8)APS: 10.5 -> 19.3Improved
increase LIE-T2T layers from 2 to 3 (ablation)WB-DETR (2-8) · COCO val2017 (WB-DETR 2-8)AP: 40.2 -> 40.3No clear effect
use LIE-T2T instead of T2T (ablation)WB-DETR (2-8) · COCO val2017 (WB-DETR 2-8)AP: 38.1 -> 40.2Improved
use LIE-T2T instead of T2T (ablation)WB-DETR (2-8) · COCO val2017 (WB-DETR 2-8)AP75: 41.4 -> 44.8Improved
use LIE-T2T instead of T2T (ablation)WB-DETR (2-8) · COCO val2017 (WB-DETR 2-8)AP50: 62.8 -> 63.2No clear effect
use smaller patch/step (8/4 vs 16/8) (ablation)WB-DETR (2-8) · COCO val2017 (WB-DETR 2-8)AP: 40.2 -> 41.1Improved
use larger patch/step (32/16 vs 16/8) (ablation)WB-DETR (2-8) · COCO val2017 (WB-DETR 2-8)AP: 40.2 -> 36.3Hurt
use non-overlapping patches (16/16 vs overlapping 16/8) (ablation)WB-DETR (2-8) · COCO val2017 (WB-DETR 2-8)AP: 40.2 -> 35.4Hurt
use non-overlapping 8/8 vs overlapping 16/8 (ablation)WB-DETR (2-8) · COCO val2017 (WB-DETR 2-8)AP: 40.2 -> 39.7Hurt
remove CNN backbone: WB-DETR (2-12) vs DETR (ablation)WB-DETR (2-12) · COCO val2017AP: DETR 42.0 vs WB-DETR (2-12) 41.8No clear effect
remove CNN backbone: WB-DETR (2-12) vs DETRWB-DETR (2-12) · COCO val2017FPS: 28 -> 36Improved

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