AI Paper Atlas

Research map / Detection and segmentation / CNN features

Res2Net: A New Multi-Scale Backbone Architecture

Shanghua Gao, Ming‐Ming Cheng, Kai Zhao, Xinyu Zhang, Ming–Hsuan Yang, Philip H. S. Torr

TPAMI 2019 · Detection and segmentation › CNN features · approach: Convolution-transformer backbones · 3,517 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
replace bottleneck block with Res2Net module (s=4)ResNet-50 · ImageNet valtop-1 error (%): 23.85 -> 22.01Improved
replace block with Res2Net moduleResNeXt-50 · ImageNet valtop-1 error (%): 22.61 -> 21.76Improved
replace block with Res2Net moduleDLA-60 · ImageNet valtop-1 error (%): 23.32 -> 21.53Improved
add Res2Net module on top of bLResNetbLResNet-50 · ImageNet valtop-1 error (%): 22.41 -> 21.68Improved
replace block with Res2Net module in 101-layer networkResNet-101 · ImageNet val (deeper networks)top-1 error (%): 22.63 -> 20.81Improved
increase scale from 2 to 8 at preserved complexity (ablation)Res2Net-50 · ImageNet val (scale ablation)top-1 error (%): 22.68 -> 22.01 -> 21.86Improved
increase scale from 4 to 8 with increased complexity (ablation)Res2Net-50 · ImageNet val (scale ablation)top-1 error (%): 22.01 -> 21.42 -> 20.80Improved
use narrower 18w x 4s Res2Net-50-L (ablation)Res2Net-50 · ImageNet val (scale ablation)top-1 error (%): 23.85 -> 22.92 with 2.9G FLOPs vs 4.2GImproved
increase scale beyond 4 on small images (ablation)Res2NeXt-29 · CIFAR-100top-1 error (%)No clear effect
add SE block to Res2NeXt-29 (ablation)Res2NeXt-29 6c x 24w x 6s · CIFAR-100top-1 error (%): 16.79 -> 16.56Improved
replace backbone with Res2Net-50Faster R-CNN · COCO detectionAP (%): 31.1 -> 33.7Improved
replace backbone with Res2Net-50DeepLab v3+ · PASCAL VOC12 valmean IoU (%): 77.7 -> 79.2Improved

Later papers that compared with this method

1 later accepted paper reports Res2Net-101 as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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