AI Paper Atlas

Research map / Detection and segmentation / CNN features

Representation Sharing for Fast Object Detector Search and Beyond

Yujie Zhong, Zelu Deng, Sheng Guo, Matthew R. Scott, Weilin Huang

ECCV 2020 · Detection and segmentation › CNN features · approach: Region-based and query-based detectors · 16 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
use the proposed diverse transformations instead of DARTS operations (ablation)FCOS searched head · COCO minivalAP: 39.0 -> 40.3Improved
add transformations with RFs up to 9 (full set vs RFs {3,5}) (ablation)searched FCOS head · COCO minivalAP: 39.2 -> 40.3Improved
allow both standard and depthwise separable conv (vs standard only) (ablation)searched FCOS head · COCO minivalAP: 39.9 -> 40.3Improved
allow both conv types (vs depthwise separable only) (ablation)searched FCOS head · COCO minivalAP: 39.5 -> 40.3Improved
add decoupling in representation sharing (ablation)searched FCOS head · COCO minivalAP: 40.0 -> 40.3Improved
add representation sharing (ablation)search procedure · COCO minivalsearch cost (GPU-days): 2.3 -> 0.6Improved
search on COCO instead of VOCsearch procedure · COCO minivalAP: 40.3 -> 40.3No clear effect
use sequential classification/regression groups instead of parallel groups (ablation)searched FCOS head · COCO minivalAP: 40.0 -> 40.3Improved
reverse the order of sequential groups (ablation)searched FCOS head · COCO minivalAP: drop of 0.1 APHurt
replace FCOS head with searched moduleFCOS · COCO minival (MobileNetV2)AP: 31.3 -> 32.7Improved
replace RetinaNet head with searched moduleRetinaNet ResNeXt-101 · COCO minivalAP: 39.8 -> 41.6Improved
replace mask head with searched cellMask R-CNN ResNet-50 · COCO minivalmask AP: 34.2 -> 35.5Improved

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