AI Paper Atlas

Research map / Detection and segmentation / CNN features

FNA++: Fast Network Adaptation via Parameter Remapping and Architecture Search

Jiemin Fang, Yuzhu Sun, Qian Zhang, Kangjian Peng, Yuan Li, Wenyu Liu, Xinggang Wang

TPAMI 2020 · Detection and segmentation › CNN features · approach: Efficient DETR query design · 73 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 FNA++ adaptation to MobileNetV2MobileNetV2 seed · Cityscapes DeepLabv3mIOU: 75.5 -> 77.0Improved
add FNA++ adaptation to MobileNetV2MobileNetV2 seed · MS-COCO RetinaNetmAP: 32.8 -> 34.7Improved
add FNA++ adaptation to MobileNetV2MobileNetV2 seed · MPII SimpleBaselinePCKh@0.5: 85.9 -> 87.0Improved
add FNA++ adaptation to ResNet-50ResNet-50 · MS-COCO RetinaNetmAP: 35.5 -> 36.8Improved
add FNA++ adaptation to FBNet-CFBNet-C · MS-COCO RetinaNetmAP: 34.9 -> 35.5Improved
remap parameters from the super network instead of the seed network (ablation)FNA++ · Cityscapes DeepLabv3mIOU: 76.6 -> 77.1Improved
remap from super network instead of seed network (ablation)FNA++ · MS-COCO RetinaNetmAP: 33.7 -> 34.7Improved
remap from super network instead of seed with 2x schedule (ablation)FNA++ · MS-COCO RetinaNet 2xmAP: 35.6 -> 36.0Improved
remap from super network instead of seed network (ablation)FNA++ · MS-COCO SSDLitemAP: 24.0 -> 24.0No clear effect
add parameter remapping before architecture adaptation (ablation)FNA++ · Cityscapes DeepLabv3mIOU: 76.0 -> 76.6Improved
add parameter remapping before parameter adaptation (ablation)FNA++ · Cityscapes DeepLabv3mIOU: 73.0 -> 76.6Improved
replace ImageNet pre-training of target architecture with remapping (ablation)FNA · Cityscapes DeepLabv3mIOU: 76.5 -> 76.6Improved

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