AI Paper Atlas

Research map / Architectures and efficiency / CNN design

InternImage: Exploring Large-Scale Vision Foundation Models With Deformable Convolutions

Wenhai Wang, Jifeng Dai, Zhe Chen, Zhenhang Huang, Zhiqi Li, Xizhou Zhu, Xiaowei Hu, Tong Lu, Lewei Lu, Hongsheng Li, Xiaogang Wang, Yu Qiao

CVPR 2023 · Architectures and efficiency › CNN design · approach: Efficient CNN scaling and dynamic inference · 982 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 multi-group spatial aggregation (ablation)InternImage-T / DCNv3 · ImageNet-1K valtop-1 accuracy: 82.3 -> 83.5Improved
Add multi-group spatial aggregation (ablation)InternImage-T / DCNv3 · COCO val2017, Mask R-CNN 1×box AP: 43.8 -> 47.2Improved
Normalize modulation scalars with softmax across sampling points (ablation)InternImage-T / DCNv3 · ImageNet-1K valtop-1 accuracy: 65.7 -> 83.5Improved
Normalize modulation scalars with softmax across sampling points (ablation)InternImage-T / DCNv3 · COCO val2017, Mask R-CNN 1×box AP: 38.7 -> 47.2Improved
Share weights among convolution neurons (ablation)InternImage-T / DCNv3 · ImageNet-1K valtop-1 accuracy: 83.6 -> 83.5No clear effect
Share weights among convolution neurons (ablation)InternImage-T / DCNv3 · COCO val2017, Mask R-CNN 1×box AP: 47.4 -> 47.2No clear effect
Increase DCNv3 kernel size from 3×3 to 5×5 (ablation)InternImage-T · ImageNet-1K valtop-1 accuracy: 83.5 -> 83.6No clear effect
Increase DCNv3 kernel size from 3×3 to 7×7 (ablation)InternImage-T · ImageNet-1K valtop-1 accuracy: 83.5 -> 82.8Hurt
Use scaling factors α=1.03 and β=1.40 instead of α=1.09 and β=1.36 (ablation)InternImage-T scaled to about 100M parameters · ImageNet-1K valtop-1 accuracy: 84.9 -> 84.5Hurt
Use scaling factors α=1.12 and β=1.34 instead of α=1.09 and β=1.36 (ablation)InternImage-T scaled to about 100M parameters · ImageNet-1K valtop-1 accuracy: 84.9 -> 83.1Hurt
Use a 3×3 dynamic sparse deformable convolution instead of dense large-kernel convolution (ablation)InternImage · ImageNet-1K val, 224²top-1 accuracy: 83.5 vs 82.8Improved

Later papers that compared with this method

1 later accepted paper reports InternImage-XL 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