AI Paper Atlas

Research map / Detection and segmentation / CNN features

Scale-Aware Modulation Meet Transformer

Weifeng Lin, Ziheng Wu, Jiayu Chen, Jun Huang, Lianwen Jin

ICCV 2023 · Detection and segmentation › CNN features · approach: Plain and high-resolution ViTs · 238 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-Head Mixed Convolution in place of single-head convolution (ablation)attention-free SMT-T · ImageNet-1K valtop-1 accuracy: 80.0 -> 80.8Improved
add Scale-Aware Aggregation in place of single linear layer (ablation)SMT-T with MHMC · ImageNet-1K valtop-1 accuracy: 80.8 -> 81.6Improved
add Evolutionary Hybrid Network (attention in last two stages, hybrid stacking) (ablation)SMT-T with MHMC and SAA · ImageNet-1K valtop-1 accuracy: 81.6 -> 82.2Improved
use 4 MHMC heads instead of 1 (ablation)SMT-T · ImageNet-1K valtop-1 accuracy: 81.8 -> 82.2Improved
increase MHMC heads beyond 4 (6 or 8) (ablation)SMT-T · ImageNet-1K valtop-1 accuracy: 82.2 -> 81.9 (6 heads), 82.0 (8 heads)No clear effect
more MHMC heads lowers throughput (ablation)SMT-T · ImageNet-1K val (V100)throughput (images/s): 983 -> 833 (1 to 4 heads)Hurt
use SAA instead of no aggregation (ablation)SMT-T · ImageNet-1K valtop-1 accuracy: 81.5 -> 82.2Improved
use SAA instead of IBN aggregation (ablation)SMT-T · ImageNet-1K valtop-1 accuracy: 82.1 -> 82.2 with 12.1M -> 11.5M paramsImproved
alternate SAM and MSA blocks in penultimate stage rather than SAM-first-half then MSA-second-half (ablation)SMT-T · ImageNet-1K valtop-1 accuracy: 81.9 -> 82.2Improved
use hybrid SAM+MSA stacking instead of all-SAM (ablation)SMT-T · ImageNet-1K valtop-1 accuracy: 81.4 -> 82.2Improved
use hybrid stacking instead of all-MSA stacking (ablation)SMT-T · ImageNet-1K valtop-1 accuracy: 81.8 -> 82.2Improved
SMT-B backbone vs Swin-B in Mask R-CNN 1xMask R-CNN · COCO valbox AP: 46.9 -> 49.0Improved

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