AI Paper Atlas

Research map / Detection and segmentation / CNN features

DQ-DETR: DETR with Dynamic Query for Tiny Object Detection

Yi-Xin Huang, Hou-I Liu, Hong-Han Shuai, Wen-Huang Cheng

ECCV 2024 · Detection and segmentation › CNN features · approach: Efficient DETR query design · 124 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 categorical counting, dynamic query selection and feature enhancement to DINO-DETR (ablation)DINO-DETR · AI-TOD-V2 testAP: 25.9 -> 30.2Improved
add counting-guided feature enhancement to counting plus dynamic queries (ablation)DQ-DETR · AI-TOD-V2 testAP: 28.1 -> 30.2Improved
add dynamic query selection to counting plus feature enhancement (ablation)DQ-DETR · AI-TOD-V2 testAP: 29.1 -> 30.2Improved
add counting and dynamic queries without feature enhancement (ablation)DINO-DETR · AI-TOD-V2 testAP_vt: 12.7 -> 12.3Hurt
regress instance count instead of classifying into four levels (ablation)categorical counting module · AI-TOD-V2 testAP: 30.2 -> 14.9Hurt
split counts into five classes instead of four (ablation)categorical counting module · AI-TOD-V2 testAP: 30.2 -> 28.8Hurt
use dynamic number of queries instead of fixed 900 for images with N <= 10 (ablation)DQ-DETR vs DINO-DETR · AI-TOD-V2, N <= 10AP: 22.5 -> 26.1Improved
use 1500 queries for very dense images instead of fixed 900 (ablation)DQ-DETR vs DINO-DETR · AI-TOD-V2, N > 500AP: 13.5 -> 14.7Improved
apply DQ-DETR to general detection (COCO) instead of DINO-DETRDINO-DETR · COCO valAP: 51.3 -> 50.2Hurt
apply DQ-DETR instead of DINO-DETR on VisDroneDINO-DETR · VisDrone valAP: 35.8 -> 37.0Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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